Forem: Burve (Burve Story Lab)

You Are Doing AI Images Wrong

Burve (Burve Story Lab) — Wed, 25 Feb 2026 11:30:00 +0000

Introduction

To be clear, there is no wrong way to make AI art. However, more often than not, people rely on the simplest approach available and settle for whatever comes out. And that is perfectly fine—if the goal is simply to “create an image with AI.” The novelty of writing a few sentences and watching an image appear that resembles your instructions is undeniable. It feels almost magical the first time you do it, and even after hundreds of generations, there is still a certain thrill in seeing the AI interpret your words visually. But if the goal is to create a specific image—something with a particular composition, a particular character, a particular mood—then “close enough” might not be “good enough.”

The gap between a casual AI-generated image and one that looks exactly the way you imagined it is not about talent or artistic skill in the traditional sense. It is about process. The people who consistently produce stunning, precisely controlled AI images are not using some secret model that the rest of us lack access to. They are simply approaching the problem differently—with more preparation, more iteration, and more tools in their workflow.

In this article, I will outline four distinct levels of AI image creation techniques, from Beginner to Professional. That said, the Beginner approach is not inherently inferior to the Professional one. The only real difference is how you approach each problem. Each higher level builds on everything from the levels below it, adding just a few extra workflows. You do not need to jump straight to the Professional level to get great results—but understanding all four levels will help you recognize when a more involved approach is worth the effort.

Beginner: Text Prompts and Regeneration

The simplest way to create an AI image is to write a text prompt describing exactly what the image should contain. The more detail you include in the prompt, the more precise the result will be. Nearly all AI image generation models accept text as input, and the output will vary depending on the model—but it will always be roughly close to the desired idea.

The issue, of course, is that “roughly close” part. You might get lucky and produce the exact image you envisioned on the first try, or the result might be “good enough.” But if neither is the case, the simplest fix is to just generate again—and again, and again—until either your credits or your patience runs out. Eventually, you will end up with the “best of the bunch,” and that becomes your final image.

There is nothing wrong with this approach, and a few practical habits can make it more effective. First, be specific. Instead of “a woman in a forest,” try describing the lighting, the time of day, what she is wearing, where she is looking, and what the forest looks like. Second, pay attention to how different models interpret language. Some models respond better to comma-separated tags, while others prefer natural sentences. Third, experiment with style keywords. Phrases like “cinematic lighting,” “shallow depth of field,” or “35mm film photography” can dramatically shift the mood and quality of the output.

Even at this level, the key insight is that prompt writing is a skill. The more you practice and study what works, the fewer regenerations you will need to get a satisfying result. Sometimes regenerating is all it takes to land on something perfect, or at least better. Do not underestimate the power of a well-written prompt paired with a little patience.

Intermediate: Reference Images

A more advanced approach is to provide the AI model with reference materials alongside your text prompt. Not all models support this, but those that do will gladly accept one or more images together with your written description. This offers a major advantage over text alone: instead of struggling to describe a specific object in words, you can simply provide a reference image, and the AI will incorporate that exact object into the generated result.

The types of references you can use vary widely. A photograph of a real person can serve as a face or character reference. A product photo can ensure the AI renders a specific item accurately. A screenshot from a film or a painting can set the mood, color palette, or lighting style. Some models even accept depth maps or edge-detection images that define the structural composition without dictating the visual content. The more relevant information you give the AI, the less it has to guess—and the less it guesses, the closer the output will be to your vision.

One important consideration at this level is the quality of your reference images. Blurry, low-resolution, or poorly lit references will introduce noise into the generation process. The AI will try to replicate what it sees, so if the reference itself is flawed, those flaws tend to carry over. Clean, well-lit, high-resolution references consistently produce better results.

Having one or more carefully chosen reference images can dramatically simplify the process of creating the exact image you have in mind. For many use cases—character consistency across a series of illustrations, product mockups, or branded content—this level is where AI image generation starts to feel genuinely useful as a creative tool rather than a novelty.

Advanced: Full Composition Control

The next step up is to stop describing the composition to the AI and instead treat it as a tool for combining pre-prepared image elements. At the Intermediate level, you provide a reference or two and let the AI handle the rest of the scene. At the Advanced level, you prepare nearly every aspect of the image in advance and ask the AI to assemble it.

For example, if the image must feature a specific woman wearing a specific outfit in a specific setting and pose, you can achieve that by providing an image of the location, a character sheet of the woman, and even a rough sketch or stand-in reference for the pose. The AI will then combine all of these elements into a single, coherent image. You are no longer hoping the AI will interpret your words correctly—you are showing it exactly what you want.

One practical example: take a photo of a dollhouse with a doll positioned in the desired pose. Pair that with a character sheet showing the character from multiple angles and a style reference image that defines the visual aesthetic you are after. The AI will use the dollhouse photo for spatial composition and pose, the character sheet for the person’s appearance, and the style reference for the overall look and feel. The result is a highly controlled image that would have been nearly impossible to achieve through text prompting alone.

This level does require more preparation. You might need to create rough sketches, find or build physical mockups, or generate preliminary AI images that serve as compositional guides for the final generation. It can feel like a lot of work upfront, but the payoff is significant: instead of generating dozens of images and hoping one lands close to your vision, you often get what you need within just a few attempts. At this level, you are essentially art-directing the AI rather than just prompting it.

Professional: Post-Processing and Targeted Editing

The leap to professional-level image generation has less to do with prompting and more to do with the extra tools you bring to the table and how you use them. No matter how carefully you prepare your references and prompts, the AI will probably still get something wrong. A hand might look slightly off. The eyes might not quite match the reference. A background element might be distracting. At every level below this one, the default response to such problems is to regenerate and hope for a better roll of the dice. At the Professional level, you fix it.

With access to modern image editing software—Adobe Photoshop, GIMP, Affinity Photo, or even free browser-based editors—you can extract the problematic part of the image, process it through the AI separately, and then place the corrected piece back into the original. The workflow is straightforward: select the area that needs fixing, cut or copy it out, run it through an AI generation or inpainting pass with appropriate prompts and references, and composite the result back into the main image. Most editing software makes this kind of compositing quick and painless once you are familiar with the basics.

This approach does not just help with small clean-ups. It also enhances detail in a way that is unique to AI-assisted workflows. No matter how small the area you cut out for correction, the AI will return a relatively large result—most models output at a fixed resolution regardless of how little source material you provide. When you scale that corrected piece back down to fit the original image, the details will be noticeably sharper and more refined than what was there before. This makes the technique especially valuable for faces, hands, text, fine textures, and any other area where detail matters.

Professional-level workflows often involve multiple rounds of this process. You might fix the hands first, then the background, then make a final pass to adjust lighting consistency across the composited areas. Each round brings the image closer to perfection. It takes more time, but the result is an image that looks intentional and polished rather than “obviously AI.”

A Note on Reference Materials

Regardless of which level you are working at, the type of reference material you provide can make or break your results. A simple frontal photo of a subject is usually enough to place that person in a scene. But sometimes a person looks different from behind, or their outfit has distinct details visible only from certain angles. In those cases, a reference sheet showing the front, side, and back views—along with a close-up of the face—will help the AI maintain a much more consistent look across different compositions and camera angles.

The same principle applies to objects, environments, and even artistic styles. If you want the AI to faithfully reproduce a specific car, a single photo from one angle might lead to creative guesses about what the other side looks like. Multiple angles eliminate that guesswork. If you want a consistent illustration style across a series of images, providing several examples of that style—rather than just one—gives the AI a much clearer understanding of what you are after.

Think of reference materials as a visual vocabulary. The richer and more specific your visual vocabulary, the more precisely you can communicate with the AI. A single reference image is a sentence. A full reference sheet with multiple angles, a style guide, and a compositional sketch is an entire brief.

Final Thoughts

All that said, even professionals will occasionally scrap everything and regenerate from scratch in hopes of faster, better results. The simple approach is always valid. There is a reason every level of this framework includes the option to just try again—sometimes the AI surprises you, and a fresh generation gives you something better than hours of careful editing would have.

The real takeaway is not that you need to adopt the most complex workflow possible. It is that you should be aware of your options. If a quick text prompt gives you exactly what you need, that is a win. If it does not, you now know that reference images, full composition control, and targeted post-processing are all available to you—and each one brings you meaningfully closer to the image you originally envisioned.

In the end, as with any creative project, the more effort you put into an AI-generated image, the closer the result will be to what you had in mind. A quick prompt will give you something. But with more thought about the process and the outcome, you can get much closer to exactly what you need. The tools are there. The techniques are there. The only question is how far you want to take it. Start wherever you are comfortable, experiment with the next level when you are ready, and remember that every great AI image, no matter how polished the final result, started with someone simply typing a prompt and pressing generate.

Putting It All Together - A LaTeX Template for Fiction Writers

Burve (Burve Story Lab) — Mon, 02 Feb 2026 11:30:00 +0000

Introduction

Over the past few articles, I've explored different aspects of writing with technology: using AI as an assistant rather than a replacement, leveraging developer tools like Git for version control, and harnessing LaTeX's custom commands for consistent formatting. Each piece addressed a specific problem, but they remained separate ideas—useful individually, yet not quite forming a complete workflow.

This article brings everything together.

I've created a LaTeX template specifically designed for fiction writers—one that incorporates the approaches I've discussed while remaining accessible to authors who've never touched LaTeX before. The template solves a problem I struggled with personally, and I'm sharing it because I suspect other writers face the same frustrations.

The Problem I Needed to Solve

When I first started using AI assistance for editing—the approach I described in my first article—I made a mistake that seemed minor at the time. I would write a passage, send it to AI for polish, and then paste the edited version directly over my original text. Why keep the rough draft? I reasoned that the AI chat history preserved everything. If I ever needed to recover my original writing, I could find it there.

That reasoning collapsed the first time I actually needed to go back.

Searching through dozens of chat sessions, trying to remember which conversation contained which chapter, scrolling through walls of back-and-forth exchanges looking for a specific paragraph—it was tedious, frustrating, and sometimes impossible. Chat histories aren't organized by chapter or scene. They're organized by when you happened to have the conversation, which has nothing to do with how your manuscript is structured.

I needed a better system. I needed to preserve my original writing in a way that kept it connected to the edited version, so I could compare them, switch between them, or recover the original if an edit went wrong.

At the same time, I was battling Microsoft Word. I've never made peace with Word's formatting. Margins that shift inexplicably. Styles that don't apply consistently. Headers that decide to renumber themselves. Every manuscript became a war between me and the software, and the software usually won. I'd spend hours fixing formatting issues that had nothing to do with the actual writing.

LaTeX solved the formatting problem completely. The PDFs it produces are professionally typeset—consistent margins, proper typography, clean chapter headings. More importantly, the formatting just works. I define it once and forget about it. No more battles.

But I still needed to solve the original-versus-edited problem within LaTeX. That's where the template came from.

The Dual-File Workflow

The core innovation in this template is simple: every chapter part exists in two versions, stored in separate folders.

The Original folder contains your raw writing—the first draft, exactly as it came out of your head. Rough, unpolished, probably riddled with passive voice and inconsistent punctuation. This is your authentic creative output, preserved permanently.

The Edited folder contains the polished version—whether you edited it yourself, worked with a human editor, or used AI assistance. The story and voice remain the same; the execution is refined.

When you compile the manuscript into a PDF, the template automatically checks which version to use. If an edited version exists, it uses that. If not, it falls back to the original. You can also override this behavior globally—one setting in the configuration file switches the entire manuscript between original and edited versions.

Why does this matter?

First, you never lose your original writing. It's always there, in a clearly labeled location, exactly as you wrote it. No digging through chat histories. No wondering which version is which.

Second, you can compare versions easily. Open both files side by side and see exactly what changed. This is invaluable for learning from the editing process—understanding what AI or a human editor improved helps you write better first drafts over time.

Third, you can revert selectively. If an edit went too far—if the polished version lost something important from the original—you can pull specific passages back. The original is always available as a reference.

Fourth, the workflow integrates naturally with Git version control. Both your original and edited folders are tracked, giving you complete history of how your manuscript evolved. You get the benefits I described in my article on developer tools, applied to a structure that makes sense for fiction writing.

What the Template Includes

Beyond the dual-file workflow, the template provides several features designed to make LaTeX accessible to authors who aren't technically inclined.

Page Format Options

One line in the settings file controls your entire page layout. Choose from A5 novel size (the European standard for fiction), A4 for drafts and manuscripts, US Letter for American printing, 6×9 trade paperback, or 5×8 pocket book format. Change that single line, recompile, and your entire manuscript reformats itself. No manual adjustment required.

Character Name Commands

As I discussed in my LaTeX article, custom commands let you define character names once and use shortcuts throughout your manuscript. The template includes a dedicated file for these definitions. Instead of typing "Lord Valdermort the Destroyer" every time your antagonist appears, you type \villain and the full name appears in the compiled PDF.

This isn't just convenient—it's protective. You'll never misspell a character name because you're not typing it. And if you decide mid-draft that "Valdermort" is too close to a certain famous dark wizard, you change the definition once and every instance updates automatically.

Built-In Dialogue and Scene Commands

The template includes pre-built commands for common fiction elements:

\say{Alice}{Hello, how are you?} — formats character dialogue consistently

\say{Bob}[whispering]{I'm not sure about this.} — dialogue with an optional modifier

\thought{Alice}{What is he hiding?} — internal thoughts, styled distinctly

\sceneBreak — a visual scene separator

\sfx{BOOM!} — sound effects with appropriate styling

These commands ensure visual consistency across your manuscript. Every piece of dialogue looks the same. Every scene break uses identical formatting. You focus on the story; the template handles presentation.

Professional Output

The template produces PDFs with drop caps at chapter openings, proper typography throughout, and a clickable table of contents. These are small details, but they add up to a document you can share without embarrassment—whether with beta readers, potential agents, or directly to readers if you're self-publishing.

The Character Profile System

The template includes a character profile template—a structured checklist for documenting everything about your characters. This isn't a LaTeX feature specifically; it's a reference document that lives alongside your manuscript.

The profile covers basic information (name, age, role), physical characteristics, background and history, motivations and goals, personality traits, speech patterns with example dialogue, special abilities, character arc planning, relationships, and detailed visual descriptions for illustration purposes.

You can fill these profiles manually, treating them as a structured way to think through your characters before and during writing. Alternatively, you can use AI to help generate or expand character details—paste the template into a conversation and ask AI to help flesh out a character based on your initial concept.

The profile template also includes the LaTeX command definitions for each character, keeping everything in one place. When you create a new character, you document their details and define their name commands in the same file, then copy the commands to the main character definitions file.

One note on the template's design: it works entirely without AI if you prefer. Nothing in the template requires AI assistance. The structure is useful regardless of how you choose to fill it out. I've designed it this way deliberately—some authors want to embrace AI tools, others prefer to maintain distance, and the template should serve both approaches.

Getting Started

The template is available on GitHub: [https://github.com/Burve/CreativeWritingLaTeXFramework\]

To use it, you'll need a LaTeX distribution installed on your computer. On Windows, MiKTeX is the most common choice. On Mac, MacTeX. On Linux, TeX Live. All are free. You'll also want a text editor—I recommend Visual Studio Code with the LaTeX Workshop extension, which provides syntax highlighting and one-click PDF compilation.

The repository README includes detailed setup instructions, but the basic workflow is straightforward:

First, fork or download the template from GitHub.

Second, edit the settings file with your book's title, author name, and preferred page format.

Third, add your character name commands to the characters file.

Fourth, create your chapter content in the Original folder, using the provided chapter files as templates.

Fifth, compile the main.tex file to generate your PDF.

When you're ready to edit, copy your original files to the Edited folder, make your changes (or use AI assistance), and recompile. The template handles the rest.

Honest Limitations

This template isn't for everyone, and I should be clear about its constraints.

If you're submitting to traditional publishers or agents who require Word documents, you'll need to convert your LaTeX output. Tools like Pandoc can handle this, but it's an extra step, and complex formatting may need manual adjustment. The template works best for self-publishing workflows or for authors who want a polished drafting environment and can convert for submission later.

LaTeX has a learning curve. The template minimizes this by pre-configuring most settings, but you'll still need to understand basic LaTeX syntax—how to use commands, how files relate to each other, how compilation works. Expect a few hours of orientation before things feel comfortable.

Collaboration with non-LaTeX users requires workflow adjustments. If your editor works in Word, you'll need to export, receive their changes, and manually incorporate them back into your LaTeX files. This is manageable but not seamless.

The template provides structure, not magic. You still need to write the book. You still need to develop characters, craft plots, and do the hard creative work that no tool can automate. What the template offers is a framework that stays out of your way while keeping your work organized.

Going Further

The template I've shared provides everything you need to start writing fiction in LaTeX with a sensible original/edited workflow. It includes the basic commands, the file structure, and the configuration options to produce professional-quality PDFs.

But it's intentionally a starting point, not an endpoint.

I've developed additional custom commands for my own writing—specialized formatting for flashbacks, letters, different narrative modes, and various stylistic elements. I've also developed prompts for using AI assistance effectively with this workflow, including structured approaches for character development, dialogue refinement, and consistent voice editing.

If you want access to those additional resources, ready-made prompts, and a community of authors using this approach to share techniques and solve problems together, I've created a space for that: [https://www.skool.com/burve-story-lab-5890 ]

The community includes expanded command libraries, AI prompt templates for various writing tasks, and direct access to ask questions and get help with LaTeX issues. It's a paid community because maintaining it takes time, and I want to ensure everyone there is genuinely invested in the approach.

That said, the free template on GitHub is fully functional. You don't need the community to benefit from the dual-file workflow, the character commands, or the professional formatting. Everything I've described in this article works out of the box. The community is for authors who want to go deeper—more commands, more prompts, more collaboration with like-minded writers.

Conclusion

This series has been about finding better ways to write—not replacing the creative work, but removing friction from everything around it. AI assistance handles polish without taking over your voice. Version control protects your work while enabling fearless experimentation. LaTeX commands automate consistency so you can focus on story. And now, a template that brings these pieces together into a coherent workflow.

The tools don't make you a better writer. Practice does. Reading widely does. Putting in the hours does. But the right tools can make the process less frustrating, more organized, and ultimately more sustainable. They can help you spend more time on the creative work that matters and less time fighting with software that doesn't care about your story.

I built this template because I needed it. I'm sharing it because I suspect you might need it too.

Happy writing.

LaTeX for Creative Writing: A Fiction Author's Secret Weapon

Burve (Burve Story Lab) — Mon, 19 Jan 2026 14:20:00 +0000

Introduction

When writers think about LaTeX, they think about academic papers. Dense research documents filled with equations, citations, and footnotes. They don't think about fantasy novels, thriller manuscripts, or literary fiction.

That's a missed opportunity.

LaTeX—the typesetting system beloved by mathematicians and scientists—offers creative writers something word processors can't match: programmable consistency. The same features that let physicists define custom notation for quantum equations let novelists define custom styling for character dialogue, narrative voices, and recurring textual elements. Once you see what's possible, you might wonder why more fiction authors haven't discovered this tool.

What Is LaTeX, Exactly?

If you're unfamiliar with LaTeX (pronounced "LAH-tek" or "LAY-tek"), here's a quick orientation. Unlike Microsoft Word or Google Docs, where you see your document as it will appear while you write, LaTeX separates content from presentation. You write in a plain text file with markup commands, then "compile" that file into a beautifully formatted PDF.

Think of it like HTML for print documents. You write \textbf{bold text} instead of clicking a bold button, and \textit{italic text} instead of pressing Ctrl+I. This might sound like extra work—and initially, it is. But the payoff comes from what this approach enables: custom commands that automate repetitive formatting across an entire manuscript.

The PDF files LaTeX produces aren't just "good enough"—they're professionally typeset. LaTeX's algorithms handle kerning, ligatures, and hyphenation with sophistication that consumer word processors can't match. The difference is subtle but cumulative: text that's easier to read, margins that feel balanced, typography that signals quality before anyone reads a word.

The Power of Custom Commands

Here's where LaTeX becomes genuinely exciting for fiction writers: custom commands.

Imagine you're writing a novel with an omniscient narrator whose voice appears in italics, distinguished from the close third-person perspective of your main chapters. In Word, you'd select each narrator passage and apply italic formatting manually. If you later decide the narrator's voice should be in a different font rather than italics, you'd need to find every single passage and change it—a tedious, error-prone process in a full-length manuscript.

In LaTeX, you define a command once:

\newcommand{\narrator}[1]{\textit{#1}}

Then throughout your manuscript, you write:

\narrator{The town had seen better days, though no one alive could remember them.}

Every narrator passage uses the same command. If you decide midway through drafting—or during revision, or after feedback from beta readers—that the narrator should use a serif font instead of italics, you change the definition once:

\newcommand{\narrator}[1]{\begingroup\fontfamily{ptm}\selectfont #1\endgroup}

Recompile, and every narrator passage in your entire manuscript updates instantly. No hunting, no missed instances, no inconsistencies.

Practical Applications for Fiction

Character Names and Spelling Consistency

Fantasy and science fiction authors face a particular challenge: complex character and location names that are easy to misspell. Is it "Vaeloria" or "Vealoria"? "Khal'thros" or "Khalthros"? Across a 100,000-word manuscript, maintaining perfect consistency is exhausting.

LaTeX offers an elegant solution. Define your names as commands:

\newcommand{\elfgirl}{Vaeloria}

\newcommand{\darkfortress}{Khal'thros}

\newcommand{\magicsword}{Dawnbreaker}

Now you write \elfgirl instead of "Vaeloria" throughout your manuscript. You'll never misspell it because you're not typing it—the command handles the actual name. If you decide during revision that "Vaeloria" should become "Vaeleryn," you change the definition once and every instance updates.

This approach also makes find-and-replace operations surgical. Searching for \elfgirl finds only the character references you defined—not fragments of other words that happen to contain the same letters.

Styled Text Elements

Many novels include styled text elements: letters, documents, text messages, dreams, flashbacks, or internal monologue. Each might have distinct formatting. In Word, you'd create paragraph styles—but those styles can't accept arguments or nest within other content easily.

LaTeX commands can handle complex formatting with parameters:

\newcommand{\textmessage}[2]{\begin{quote}\texttt{\textbf{#1:} #2}\end{quote}}

Use it like this:

\textmessage{Sarah}{Running late. Don't start without me.}

Every text message in your manuscript will be formatted identically: indented, in monospace font, with the sender's name bolded. Change the definition, change every text message at once.

Multiple Narrative Voices

Novels with multiple point-of-view characters can use custom commands to maintain distinct formatting for each voice:

\newcommand{\povmarcus}[1]{\section*{Marcus}#1}
\newcommand{\povlena}[1]{\section*{Lena}\textit{#1}}

If Marcus's chapters are in standard text and Lena's in italics, the formatting stays consistent automatically—and can be adjusted globally at any time.

Beyond Custom Commands

Custom commands are LaTeX's most immediately useful feature for creative writers, but other capabilities deserve mention.

Automatic cross-references: Label any chapter, section, or location in your manuscript, then reference it by label. If chapter numbers change during revision, references update automatically.

Superior typography: LaTeX's typesetting algorithms produce more readable text through intelligent hyphenation, proper kerning, and optimal line breaks. Readers may not consciously notice, but the cumulative effect is text that feels more polished.

Clickable navigation: With the hyperref package, LaTeX generates PDFs with clickable tables of contents, cross-references, and bookmarks—features that require manual setup in word processors.

Plain text source files: Your manuscript exists as plain text, which means it works seamlessly with version control systems like Git. Track every change, create experimental branches, maintain complete revision history—capabilities I discussed in my previous article on developer tools for writers.

AI Makes LaTeX Accessible

Here's the practical reality: you don't need to memorize LaTeX syntax to use these features effectively.

Modern AI assistants like Claude or ChatGPT can generate LaTeX custom commands from plain English descriptions. Tell the AI what you want—"I need a command that formats text messages with the sender's name in bold, the message in a gray box, and a timestamp in small text"—and it will produce working LaTeX code. You copy the command definition into your document, then use it throughout your manuscript.

This dramatically lowers the barrier to entry. You don't need to understand the intricacies of LaTeX to benefit from its power. You need to understand what you want, describe it clearly, and let AI handle the technical implementation. When something doesn't work quite right, describe the problem and ask for adjustments.

Honest Limitations

LaTeX isn't for everyone, and honesty requires acknowledging its drawbacks.

The learning curve is real. Even with AI assistance, you'll spend time learning basic LaTeX structure, understanding how to compile documents, and troubleshooting when things go wrong. Expect several hours of orientation before you're comfortable.

Collaboration can be challenging. If your editor or co-author uses Word, you'll need to convert your LaTeX file for them to review—and their tracked changes won't automatically flow back into your LaTeX source. This is a workflow friction that traditional word processors don't impose.

Traditional submission requirements persist. Most literary agents and publishers expect .doc or .docx files. While tools like Pandoc can convert LaTeX to Word format, complex custom commands may require manual adjustment. LaTeX works best for self-publishing workflows or as a drafting environment with conversion at the submission stage.

No real-time formatting preview. You write in plain text, then compile to see the result. Some writers find this separation liberating—it keeps you focused on words rather than fiddling with fonts. Others find it frustrating. Your preference likely depends on how you're wired.

Getting Started

If you're curious about trying LaTeX for creative writing, here's a practical path forward.

First, install a LaTeX distribution. On Windows, MiKTeX is popular. On Mac, MacTeX. On Linux, TeX Live. These are free and include everything you need to compile documents.

Second, choose an editor. TeXstudio provides a dedicated LaTeX environment with syntax highlighting and one-click compilation. Alternatively, Visual Studio Code with the LaTeX Workshop extension works well and integrates with other development tools.

Third, start with a template. Don't build a manuscript structure from scratch. Find a fiction manuscript template online or use a minimal starting document, then modify it for your needs.

Fourth, define your first custom command. Pick something simple—a character name you use frequently or a basic formatting style. Write a few pages using the command. Experience the workflow before committing to an entire manuscript.

Finally, use AI as your LaTeX tutor. When you want a custom command but don't know the syntax, ask. When compilation fails, paste the error message and ask for help. The learning curve flattens dramatically when you have an always-available teacher.

Conclusion

LaTeX occupies a strange position in the writer's toolkit—familiar to academics, invisible to most fiction authors. That invisibility is understandable. The tool's reputation as technical and specialized discourages exploration. Why would a novelist use software designed for scientific papers?

The answer is custom commands: the ability to define once and apply everywhere, to maintain perfect consistency across hundreds of pages, to change your mind about formatting and implement that change in seconds rather than hours. For writers managing complex manuscripts with multiple voices, intricate naming conventions, or distinctive styled elements, these capabilities solve real problems.

Not every writer needs this. If you're happy with Word or Scrivener, if your manuscripts don't involve complex formatting requirements, there's no reason to switch. The learning investment wouldn't pay off.

But if you've ever lost hours to find-and-replace operations, if you've ever struggled to maintain consistent formatting across a long manuscript, if you've ever wished you could automate the tedious parts of document preparation—LaTeX might deserve a closer look. The tool doesn't care that you're writing fiction instead of physics. It just sees text, and gives you remarkably precise control over how that text appears.

Sometimes the best tools for creative work come from unexpected places.

Writing a Novel the Developer Way

Burve (Burve Story Lab) — Mon, 12 Jan 2026 14:20:00 +0000

Introduction

The tools you use shape how you work. Painters choose between oils and acrylics. Musicians select instruments that match their style. Writers, too, make choices about their workspace—and those choices matter more than most realize.

Most authors default to familiar options: Microsoft Word, Google Docs, or dedicated writing software like Scrivener. These tools work well and have served writers for decades. But there’s another category of software—one designed for an entirely different profession—that offers surprising advantages for creative writers willing to look beyond the obvious choices.

I’m talking about the software that developers use every day: Integrated Development Environments, or IDEs. Before you dismiss this as irrelevant technical nonsense, consider what you actually do when you write. You type letters and symbols into a file. You organize those files into folders. You revise, rewrite, and track changes over time. Developers do exactly the same thing—just with code instead of prose. The tools they’ve built to manage their work translate remarkably well to managing yours.

What Exactly Is an IDE?

An IDE—Integrated Development Environment—is software designed to help developers write, edit, and manage code. Think of it as a text editor with superpowers. Popular options include Visual Studio Code (free, from Microsoft), Cursor (built specifically around AI assistance), and Sublime Text.

“But I’m not a developer,” you might say. “Why would I use their tools?”

Here’s the thing: an IDE doesn’t care what you’re writing. It sees text. Whether that text forms a Python script or a fantasy novel, the IDE handles it the same way. You’re not learning to code by using these tools—you’re simply accessing features that word processors don’t offer.

The real question isn’t whether you can use developer tools for writing. It’s whether doing so provides benefits that traditional writing software can’t match. The answer, for many authors, is yes.

Version Control: Your Manuscript’s Safety Net

The most compelling reason to adopt developer tools is version control—specifically, a system called Git and its most popular hosting platform, GitHub.

Think of version control as a time machine for your manuscript. Every time you save a meaningful change, you create a snapshot. Days, weeks, or months later, you can return to any of those snapshots instantly. Unlike Word’s version history or Google Docs’ revision timeline, Git preserves everything—not just recent changes, and not subject to arbitrary time limits.

But version control offers more than just backup. Imagine you’re halfway through your novel when a wild idea strikes: what if the protagonist’s mentor was actually the antagonist all along? In traditional writing software, exploring this idea means either making irreversible changes or manually saving a copy of your entire project. With Git, you create what’s called a “branch”—a parallel version of your manuscript where you can experiment freely without affecting your main work.

If the experiment succeeds, you merge the branch back into your main manuscript. If it fails, you simply delete the branch. No harm done. Your original work remains untouched throughout.

Consider the practical applications:

Alternate storylines: Branch your novel to explore different plot directions without losing your original path.
Risky revisions: Test major structural changes in isolation before committing to them.
Beta reader feedback: Create separate branches for each reader’s suggested changes, then selectively merge what works.
Collaboration: Multiple authors can work on the same manuscript simultaneously, with Git intelligently merging everyone’s contributions.

GitHub adds another layer of value. Your manuscript lives in the cloud, accessible from any computer, automatically backed up, and protected against local hardware failures. You can make your repository private (for unpublished work) or public (for open-source projects or community collaboration).

The learning curve for Git basics is modest—a few hours of tutorials will teach you everything a writer needs. The payoff is a level of manuscript control that traditional writing software simply cannot match.

AI Integration: More Than Autocomplete

Modern IDEs have embraced artificial intelligence in ways that benefit writers as much as developers.

The most visible feature is inline completion. As you type, AI predicts what might come next and offers suggestions. For developers, this means auto-completing function names and code structures. For writers, it means sentence suggestions, dialogue continuations, and phrasing alternatives.

Is this always helpful? No. Sometimes you know exactly what you want to write, and AI suggestions become distracting noise. But when you’re stuck—when the next sentence won’t come, when dialogue feels stilted, when a transition eludes you—having AI offer possibilities can break the logjam. You’re not obligated to accept any suggestion. Think of it as a brainstorming partner who occasionally has good ideas.

More powerful than inline completion is the integrated AI chat feature that most modern IDEs now include. Tools like Cursor and Visual Studio Code with extensions allow you to have conversations with AI assistants—Claude, GPT, or others—directly within your writing environment. Critically, these conversations can include your manuscript files as context.

This changes how you interact with AI assistance. Instead of copying text into a separate chat window, you can ask questions like:

“Does my protagonist’s motivation stay consistent throughout chapters 3 through 7?”
“Suggest ways to foreshadow the reveal in chapter 12 based on what I’ve written so far.”
“Review this dialogue for voice consistency with my character descriptions.”

The AI sees your actual files—your notes, your outline, your draft—and provides responses grounded in your specific project. This integration creates a more seamless workflow than constantly switching between your writing software and a separate AI tool.

File Formats and Flexibility

Traditional word processors excel at formatted documents—bold text, headers, page layouts, and the visual presentation of your work. IDEs, by contrast, work primarily with plain text.

This sounds like a limitation, and in some ways it is. You won’t be adjusting margins or selecting fonts inside an IDE. But plain text formats offer advantages that formatted documents can’t match.

The most relevant format for writers is Markdown—a simple way to indicate structure (headers, bold, italic, lists) using plain text characters. A header is just a line starting with #. Bold text is wrapped in **asterisks**. The syntax is intuitive and takes minutes to learn.

Why does this matter? Markdown files work everywhere. They open in any text editor on any operating system. They convert easily to HTML for blogs, to formatted documents for print, to ebooks for digital distribution. They’re small, fast, and future-proof—you’ll never face a situation where a file format becomes obsolete or incompatible.

For authors focused on self-publishing or digital-first distribution, Markdown offers particular advantages. Blog platforms, static site generators, and many ebook creation tools accept Markdown directly. Your manuscript becomes more portable and more flexible.

A word of caution: if you’re submitting to traditional publishers or literary agents, they typically expect .doc or .docx files formatted to specific guidelines. Markdown works well for drafting and certain publishing paths, but you may need conversion tools like Pandoc for traditional submission. This is a workflow consideration, not a dealbreaker—but it’s worth understanding before you commit.

Limitations and Learning Curve

Adopting developer tools isn’t without friction. Honesty requires acknowledging the downsides.

The learning curve is real. IDEs are designed for developers, and while writers don’t need most features, the interface can initially feel overwhelming. Menus reference concepts you don’t need to understand. Features you do need aren’t always obvious. Expect a few hours of orientation before things feel comfortable.

Formatting is limited. If you need real-time visual formatting—WYSIWYG editing with visible fonts, margins, and layouts—IDEs aren’t the right choice. They’re built for content creation, not document design. Final formatting typically happens elsewhere, in a tool specifically designed for layout.

Collaboration requires buy-in. Git’s collaboration features only work when everyone uses Git. If your co-author or editor prefers Word’s Track Changes, you’ll need to accommodate that workflow rather than forcing developer tools on resistant collaborators.

Not all AI features are free. While many IDEs offer free tiers, advanced AI capabilities often require subscriptions. Cursor, for instance, offers powerful AI integration but at a monthly cost. Evaluate whether the benefits justify the expense for your specific workflow.

These limitations aren’t trivial, but neither are they insurmountable. Many writers find that the benefits—particularly robust version control and seamless AI integration—outweigh the learning investment and workflow adjustments.

Getting Started: A Practical Path

If you’re intrigued but unsure where to begin, here’s a straightforward approach.

Step 1: Download Visual Studio Code. It’s free, widely supported, and strikes a good balance between power and approachability. Install it from code.visualstudio.com.

Step 2: Create a project folder. Make a folder on your computer for your manuscript. Inside, create a file called manuscript.md (or whatever name you prefer). The .md extension indicates a Markdown file.

Step 3: Write something. Open VS Code, open your project folder, and start writing. Ignore the unfamiliar menus and features for now. Just write.

Step 4: Learn basic Markdown. Spend fifteen minutes with a Markdown tutorial. Learn headers (#), emphasis (*italic*, **bold**), and basic structure. That’s enough to start.

Step 5: Add Git when ready. Once you’re comfortable with the editor, initialize a Git repository. VS Code has built-in Git support—the Source Control panel (left sidebar) guides you through making commits. Create a free GitHub account and push your repository to the cloud.

Step 6: Explore AI features. If AI assistance interests you, install an extension like GitHub Copilot or try Cursor as your IDE. Experiment with inline suggestions and integrated chat.

Don’t try to master everything at once. Start with the basics—just using the editor as a text editor—and add capabilities incrementally as you grow comfortable.

Conclusion

Writing with developer tools isn’t about becoming technical or abandoning creative instincts for cold efficiency. It’s about recognizing that writers and developers share fundamental needs: managing text, tracking changes, exploring alternatives, and leveraging AI to work smarter.

The tools developers have built to address these needs are sophisticated, powerful, and—increasingly—accessible to non-programmers willing to learn something new. Version control protects your work while enabling fearless experimentation. AI integration meets you where you write rather than requiring constant context-switching. Plain text formats ensure your manuscripts remain portable and future-proof.

This approach won’t suit every writer. Those who need rich visual formatting, who prefer familiar word processors, or who work primarily with collaborators wedded to traditional tools may find the tradeoffs unfavorable. That’s a legitimate choice.

But for writers who value robust backup systems, who want seamless AI assistance, who publish digitally or maintain complex projects with multiple branches and versions—developer tools offer something traditional writing software can’t match.

The software doesn’t care that you’re writing a novel instead of a program. It just sees text—and helps you manage that text more effectively than you ever could before.

Perhaps it’s time to write your novel the developer way.

Writing with AI Assistance

Burve (Burve Story Lab) — Mon, 05 Jan 2026 14:20:00 +0000

Introduction

Writing with AI assistance is not the same as writing by AI.

When you write directly with AI, you're prompting it to generate content from scratch. When you write with AI assistance, you remain the author—AI simply helps refine your work rather than doing most of it for you.

The Problem with Full AI Generation

Yes, you can create a story from scratch using AI prompts, but the result will likely be short and won't capture exactly what you wanted to say.

There are workarounds for the length problem. The most common approach involves asking AI to first create a summary of a story, then expand it by dividing it into chapter summaries—for example, 12 chapters. Each chapter summary then extends into story beats, perhaps 12 beats per chapter. Finally, you can ask AI to expand each story beat to roughly 500 words. This method can yield approximately 6,000 words per chapter and 72,000 words for an entire manuscript.

Quantity-wise, you'll have a solid manuscript, and it might even have a coherent story structure. But it won't be your story if AI generates everything itself.

Depending on your goals, you could set up automation and generate one or even multiple manuscripts per day. If your aim is to mass-produce low-effort content and earn money through sheer volume, that's technically possible. But at that point, you're not really an author—you're a salesperson.

The Control Problem

Another significant issue with fully AI-generated manuscripts is the near-total loss of creative control.

Think of your story as a path on a map. If you simply give AI a starting direction, the path will wander in unpredictable ways. You can try to guide it by creating detailed story beats yourself, essentially placing pins on the map to mark where you want the path to go. But AI will find its own route between those points, and you'll have limited influence over what happens in between.

Those gaps are largely out of your hands. AI might introduce a subplot that derails your intended theme, or shift a character’s motivation in ways that contradict what comes later. By the time you notice, you’ve built subsequent chapters on a foundation you didn’t choose. Fixing it means backtracking—or accepting a story that drifted from your original vision.

You might write faster by only creating story beats, but the end result still won't be the story you want to tell. It might be close enough—but "close enough" isn't always good enough.

The AI-Assisted Approach

The alternative is writing with AI assistance. With this approach, you're in the driver's seat while AI handles the supporting work.

Imagine going on a road trip and using GPS for navigation. You're in full control of the vehicle, choosing where to stop, which route to take, and how fast to drive. But when you need help finding your way, the GPS is there to guide you. Sometimes it suggests a faster route, and you take it. Other times, you ignore its recommendation because you know a scenic detour worth making. The GPS doesn't argue—it recalculates and continues supporting your journey. That's exactly what AI-assisted writing offers: you maintain creative control while AI provides helpful support when needed.

Simply put, when you write with AI assistance, you compose the manuscript yourself, and AI helps with polishing, style consistency, voice consistency, and overall line editing. AI is almost always better at rephrasing existing content than creating something entirely new.

With fiction, you also face fewer issues with hallucinations—the tendency for AI to generate incorrect information. Since you're not creating a factual document, any invented details might actually work within your story. And for the factual elements you do include, they're easy to verify yourself.

Is It Still Your Writing?

A common concern among writers considering AI assistance is authenticity: "If AI helped me, is it still my writing?"

Consider the tools you already use. Spell-check catches your typos. A thesaurus helps you find better word choices. Grammar software flags awkward constructions. Have these tools ever made you feel like less of a writer? Probably not—because the ideas, the story, and the creative decisions remain entirely yours.

AI assistance works the same way, just with greater capability. It's a more sophisticated tool in your toolkit, not a ghostwriter. The plot you crafted, the characters you created, the themes you're exploring—none of that changes because AI helped you smooth out a paragraph or maintain consistent dialogue patterns. You're still the author. AI is simply helping you execute your vision more effectively.

Practical Applications

Creating Distinct Character Voices

One of the significant benefits of AI-assisted writing is the ability to create multiple characters with unique speech patterns. In traditional writing, you'd need to carefully track each character's voice and ensure consistency throughout. With AI assistance, you can write dialogue naturally and then instruct AI to convert specific characters' lines into their distinctive styles.

For example, consider this neutral line: "How was your trip? I was so worried about you. I hope you're all right now."

When asked to rephrase it for different character types, AI produces distinctly different results:

20-year-old valley girl:

"Oh my God, so like, how was your trip? I was literally SO worried about you—like, I couldn't even. Are you okay now? Please tell me you're okay!"

75-year-old grumpy man:

"Well? How'd it go? I wasn't gonna say anything, but I didn't sleep a wink worryin' about you. You doin' all right or what?"

35-year-old mother:

"Hey, sweetheart. How was the trip? I have to be honest—I was worried sick the whole time. Are you feeling okay now? Do you need anything?"

45-year-old Victorian gentleman:

"Pray tell, how did you find your journey? I confess I was most anxious during your absence. I do hope you have returned to us in good health and spirits."

30-year-old tech professional:

"Dude, how was the trip? Not gonna lie, I was low-key stressing about you. You good now? Everything optimized?"

15-year-old teenage boy:

"Yo, how was your trip or whatever? I mean, I wasn't like worried worried, but... yeah. You good now?"

As you can see, the core intent remains identical, but each line takes on a completely different personality. If you can already write fluently in multiple voices, this feature might not be necessary. But for most writers, it's a valuable tool for filling gaps in their range.

Line Editing and Polish

AI-assisted writing also excels at direct line editing: checking spelling, punctuation, passive voice, and other mechanical issues.

Consider what AI can catch that you might miss after staring at your manuscript for hours. It can identify passive voice constructions and suggest active alternatives—transforming "The door was opened by Sarah" into "Sarah opened the door." It can flag repetitive word usage, alerting you when you've used "suddenly" four times in two pages. It can analyze sentence rhythm, noting when you've strung together five short sentences in a row or written a paragraph of nothing but complex constructions.

AI can also help with consistency issues that are notoriously difficult to track manually. Are your dialogue tags consistent throughout? Have you accidentally switched from past to present tense mid-chapter? Does your character's name spelling stay the same? These mechanical details matter, and AI catches them without fatigue or frustration.

Of course, a human editor with years of experience will perform better—and charge accordingly. If you can afford professional editing, it's worth the investment. But AI can do a competent job, likely better than most writers can manage on their own.

Getting Started with AI-Assisted Writing

If you're ready to try AI-assisted writing, here's a straightforward workflow to begin.

First, write your draft yourself. Get your ideas down without worrying about perfection. This is your creative foundation—the story, the voice, the vision. Don't involve AI at this stage; you want the raw material to be authentically yours.

Second, identify specific tasks for AI. Rather than asking AI to "make this better," be precise. Ask it to check for passive voice. Request that it rephrase a particular character's dialogue in a gruff, working-class style. Have it identify repetitive words in a chapter. Specific prompts yield better results than vague ones.

Third, review everything. Accept the suggestions that improve your work. Reject the ones that don't fit. Modify others to better match your voice. You're the final authority—AI proposes, but you decide.

Finally, iterate. AI-assisted editing isn't a one-pass process. You might run dialogue through voice refinement, then check the whole chapter for consistency, then do a final polish for flow. Each pass serves a different purpose.

Important Caveats

AI may occasionally add extra details beyond what you've written. Sometimes these additions are welcome surprises that enhance your manuscript. Other times, they're not.

The golden rule: always read everything AI generates. Never blindly trust AI output. Review every suggestion, every edit, every addition. You can always accept what works and reject what doesn't—but only if you're paying attention.

Be especially cautious about over-reliance. If you find yourself accepting every AI suggestion without thought, you risk losing your distinctive voice. The goal is collaboration, not abdication. AI should enhance your writing, not replace your judgment.

Conclusion

AI-assisted writing represents a middle path between doing everything yourself and letting AI do everything for you. It keeps you in creative control while leveraging AI's strengths: consistency, polish, and the ability to transform content in ways that might be tedious or difficult to do manually.

Think of AI as a skilled assistant rather than a replacement author. Use it to enhance your writing, maintain consistency across characters and scenes, catch errors you might miss, and refine your prose. But remember—the story, the voice, and the vision should always be yours.

When used thoughtfully, AI assistance doesn't diminish your role as an author. It amplifies it.