Forem: Ritesh Hiremath

Most AI No-Code Tools Build Demos. I Built FoundersKit to Build Startups

Ritesh Hiremath — Fri, 02 Jan 2026 18:32:46 +0000

This is a submission for the DEV's Worldwide Show and Tell Challenge Presented by Mux

What I Built

FoundersKit is a production-ready SaaS toolkit that helps founders go from idea to a real, scalable startup in days instead of months.

In a world full of AI no-code tools that help you build fast but lock you into non-exportable, non-scalable platforms, FoundersKit focuses on something more important: ownership.

It provides founders with real, customizable code, modern SaaS architecture, and growth-ready tooling so what they launch on day one can scale into a real business.

Website: https://founderskit.in

My Pitch Video

Demo

Live Website: https://founderskit.in
How to Explore:

Visit the site to understand the value proposition and feature set
Documentation and setup guides explain how founders can quickly launch and customize their SaaS

Hehehe, a free starter template for everyone: https://github.com/Ritesh2351235/FoundersKit-starterplan

DEV judges, drop a comment on this post to get a promo code. I’m happy to offer FoundersKit Pro for free if you’re building something exciting.

The Story Behind It

Over the last year, I saw many founders, including myself, use AI no-code tools like Lovable and Bolt.new to validate ideas quickly. While these tools are impressive, they usually stop at beautiful dashboards.

The problem appears when you want to:

Export your code
Customize core logic
Scale infrastructure
Build a real startup roadmap

At that point, you realize you don’t truly own the product.

I built FoundersKit to bridge that gap, combining the speed of modern tooling with the control and scalability of real production code. The goal is simple: help founders move fast without sacrificing their future.

Technical Highlights

FoundersKit is built with real-world scalability in mind and uses modern, startup-proven technologies:

Next.js-based SaaS architecture
Pre-configured authentication and billing flows
Modular, extensible codebase for rapid customization
AI integration patterns that work in real production apps
Designed for easy deployment and long-term maintenance

Unlike no-code platforms, everything in FoundersKit is:

Fully exportable
Customizable
Deployable anywhere
Scalable without platform limitations

This makes it suitable not just for MVPs, but for long-term startup growth.

❗ By submitting this project, I confirm that my video adheres to Mux's terms of service: https://www.mux.com/terms

Thanks for checking out FoundersKit and happy building!

How I Ship SaaS Ideas in 8 Hours Instead of 8 Months

Ritesh Hiremath — Fri, 29 Aug 2025 05:37:07 +0000

I have 47 SaaS ideas sitting in my Notion.

Until last month, every single one of them started the same way: excitement for 2 days, then 6 weeks of auth hell, database schema debates, and Stripe webhook debugging. By month 3, the idea felt stale and I'd move on to the next shiny object.

Sound familiar?

Last week, I shipped 3 of those ideas. All of them are live, getting users, and generating revenue. The difference? I stopped building from scratch and started building smart.

Here's how FoundersKit turned me from a serial idea-hoarder into an actual shipping machine.

The Brutal Reality: Your Ideas Are Worthless Until They're Live

Let me be brutally honest about something every developer knows but won't admit:

We love building. We hate shipping.

Building feels productive. Setting up the perfect auth system, choosing the ideal database schema, optimizing API responses... it's comfortable. It's what we know.

But here's what I learned the hard way: In 2025, the only competitive advantage is speed.

While you're perfecting your authentication flow, someone else is already talking to users. While you're debating Next.js vs Nuxt, they're iterating based on real feedback. While you're building, they're winning.

The Idea Graveyard: My 8-Month Streak of Almost-Launches

Before FoundersKit, my development cycle looked like this:

Week 1: "This idea is genius! Let me just set up Next.js..."
Week 3: "Why is Clerk so complicated? Maybe I'll use Supabase Auth..."
Week 6: "Stripe webhooks are the devil. Let me try LemonSqueezy..."
Week 10: "I should probably build a proper landing page..."
Week 12: "This idea isn't that good anyway. I have a better one..."

Repeat. Forever.

The breaking point came when I calculated the time cost: 8 months, 6 half-built projects, 0 launched products, 0 dollars earned.

My co-founder Eshaan put it perfectly: "You're not a developer anymore. You're a professional foundation-builder who never builds the house."

That hit different.

The FoundersKit Discovery: When Everything Clicked

I discovered FoundersKit during one of my 3 AM "why is auth so hard" breakdowns. The promise seemed too good to be true:

"Launch your SaaS in hours, not months. Complete foundation + marketing system included."

Yeah, right. Another overpromised template.

But I was desperate, so I bought it. Best $88 I've ever spent.

Here's what happened next:

Hour 1: Setup That Actually Works

Downloaded FoundersKit. Followed the setup guide.

No joke - authentication, payments, database, email system, and a beautiful UI were running locally in 47 minutes.

Not "kind of working." Not "demo version." Actually working.

User registration ✅
Email verification ✅

Stripe subscriptions ✅
Dashboard with real data ✅

I stared at my screen for 10 minutes. This usually takes me 6 weeks.

Hour 3: My First Real Product Goes Live

I had this idea for a simple analytics dashboard that I'd been putting off for months. You know, something cleaner than Google Analytics but not as complex as Mixpanel.

With FoundersKit's foundation ready, I spent 2 hours building the actual unique features:

Custom event tracking
Simple, beautiful dashboards
Export functionality

By hour 3, it was live on Vercel with real Stripe payments working.

3 hours. From idea to paying customers.

The Real Magic: Non-Technical Friendly

Here's what blew my mind: my co-founder Eshaan (who's more business than code) could actually understand and modify FoundersKit.

The codebase is clean. The documentation is clear. The marketing playbooks are written in plain English.

He set up a simple feedback collection tool while I was busy with client work. A non-technical person launched a SaaS product in 8 hours.

That's when I realized FoundersKit isn't just for developers. It's for anyone with good ideas who's tired of waiting for the "technical stuff" to get sorted.

The Marketing System That Changes Everything

Every template gives you code. FoundersKit gives you customers.

The marketing playbooks include:

First 100 Users Strategy (step-by-step guide to finding your initial customers)
Product Hunt Launch Kit (I hit #4 Product of the Day following their exact plan)
Cold Outreach Templates (that don't sound like spam)
Social Media Automation (posts that actually convert)
SEO Setup (that ranks without gaming the system)

I followed their "First 100 Users" playbook for my analytics tool:

Day 1: Identified 5 indie hacker communities

Day 2: Posted helpful content (not pitches) in these communities
Day 3: Started getting signups from people who remembered my helpful posts
Week 1: 18 users, 4 paying customers

This stuff works.

The Claude Code Ultimate Guide: My Secret Weapon

Buried in FoundersKit's AI section is something called the "Claude Code Ultimate Guide."

This isn't some basic "how to use ChatGPT" tutorial. It's a complete system for using Claude to build features 10x faster.

The prompts are insane:

Generate complete feature implementations
Create database migrations
Write test suites
Build API endpoints
Generate marketing copy that converts

I used their Claude prompts to add a real-time notifications feature to my analytics dashboard. Took 40 minutes. Would have taken me 2 days before.

No more expensive courses. No more YouTube tutorials. Just working prompts that actually understand your codebase.

How I Ship SaaS Ideas in 8 Hours Instead of 8 Months (And Why Speed Is The Only Edge Left)

Ritesh Hiremath — Fri, 29 Aug 2025 05:33:55 +0000

I have 47 SaaS ideas sitting in my Notion.

Sound familiar?

Last week, I shipped 3 of those ideas. All of them are live, getting users, and generating revenue. The difference? I stopped building from scratch and started building smart.

Here's how FoundersKit turned me from a serial idea-hoarder into an actual shipping machine.

The Brutal Reality: Your Ideas Are Worthless Until They're Live

Let me be brutally honest about something every developer knows but won't admit:

We love building. We hate shipping.

Building feels productive. Setting up the perfect auth system, choosing the ideal database schema, optimizing API responses... it's comfortable. It's what we know.

But here's what I learned the hard way: In 2025, the only competitive advantage is speed.

The Idea Graveyard: My 8-Month Streak of Almost-Launches

Before FoundersKit, my development cycle looked like this:

Repeat. Forever.

The breaking point came when I calculated the time cost: 8 months, 6 half-built projects, 0 launched products, 0 dollars earned.

My co-founder Eshaan put it perfectly: "You're not a developer anymore. You're a professional foundation-builder who never builds the house."

That hit different.

The FoundersKit Discovery: When Everything Clicked

I discovered FoundersKit during one of my 3 AM "why is auth so hard" breakdowns. The promise seemed too good to be true:

"Launch your SaaS in hours, not months. Complete foundation + marketing system included."

Yeah, right. Another overpromised template.

But I was desperate, so I bought it. Best $88 I've ever spent.

Here's what happened next:

Hour 1: Setup That Actually Works

Downloaded FoundersKit. Followed the setup guide.

No joke - authentication, payments, database, email system, and a beautiful UI were running locally in 47 minutes.

Not "kind of working." Not "demo version." Actually working.

User registration ✅
Email verification ✅

Stripe subscriptions ✅
Dashboard with real data ✅

I stared at my screen for 10 minutes. This usually takes me 6 weeks.

Hour 3: My First Real Product Goes Live

I had this idea for a simple analytics dashboard that I'd been putting off for months. You know, something cleaner than Google Analytics but not as complex as Mixpanel.

With FoundersKit's foundation ready, I spent 2 hours building the actual unique features:

Custom event tracking
Simple, beautiful dashboards
Export functionality

By hour 3, it was live on Vercel with real Stripe payments working.

3 hours. From idea to paying customers.

The Real Magic: Non-Technical Friendly

Here's what blew my mind: my co-founder Eshaan (who's more business than code) could actually understand and modify FoundersKit.

The codebase is clean. The documentation is clear. The marketing playbooks are written in plain English.

He set up a simple feedback collection tool while I was busy with client work. A non-technical person launched a SaaS product in 8 hours.

That's when I realized FoundersKit isn't just for developers. It's for anyone with good ideas who's tired of waiting for the "technical stuff" to get sorted.

The Marketing System That Changes Everything

Every template gives you code. FoundersKit gives you customers.

The marketing playbooks include:

First 100 Users Strategy (step-by-step guide to finding your initial customers)
Product Hunt Launch Kit (I hit #4 Product of the Day following their exact plan)
Cold Outreach Templates (that don't sound like spam)
Social Media Automation (posts that actually convert)
SEO Setup (that ranks without gaming the system)

I followed their "First 100 Users" playbook for my analytics tool:

This stuff works.

The Claude Code Ultimate Guide: My Secret Weapon

Buried in FoundersKit's AI section is something called the "Claude Code Ultimate Guide."

This isn't some basic "how to use ChatGPT" tutorial. It's a complete system for using Claude to build features 10x faster.

The prompts are insane:

Generate complete feature implementations
Create database migrations
Write test suites
Build API endpoints
Generate marketing copy that converts

I used their Claude prompts to add a real-time notifications feature to my analytics dashboard. Took 40 minutes. Would have taken me 2 days before.

No more expensive courses. No more YouTube tutorials. Just working prompts that actually understand your codebase.

freedomFund: Your Personal Financial Literacy Partner

Ritesh Hiremath — Mon, 27 Jan 2025 05:20:19 +0000

This is a submission for the Agent.ai Challenge: Full-Stack Agent (See Details)

This is a submission for the Agent.ai Challenge: Productivity-Pro Agent (See Details)

This is a submission for the Agent.ai Challenge: Assembly of Agents (See Details)

What I Built

At FreedomFund, we believe financial literacy is a right, not a privilege. Everyone deserves to understand and take control of their finances because knowledge is the key to lasting financial independence.

Inspired by timeless principles from renowned works like Rich Dad Poor Dad and The Intelligent Investor, we’ve built a platform that simplifies and personalizes your financial journey.

Our application generates detailed, actionable financial reports tailored to your unique circumstances. From analyzing your savings and investments to providing clear visualizations of your financial health, FreedomFund equips you with the knowledge and tools to make informed decisions.

Whether you’re building wealth, planning for the future, or optimizing your finances, FreedomFund is your trusted partner, turning financial complexity into clarity and guiding you toward lasting financial freedom.

How FreedomFund Works

At FreedomFund, we aim to simplify financial literacy for everyone by turning complex financial data into actionable insights. Here's how our platform empowers you to take control of your finances:

1. Input Your Financial Data

Start by sharing details like income, expenses, savings goals, and investment preferences. The more precise your input, the more personalized and accurate your financial report will be.

2. Analyze Key Metrics

Using cutting-edge algorithms and insights from industry benchmarks, we evaluate essential metrics such as:

Savings Rate: How effectively you're saving.
Emergency Fund Coverage: Your readiness for unexpected expenses.
Investment Ratios: The balance and diversification of your investments.
Net Worth Overview: A comprehensive snapshot of your financial health.

Metric Analysis Code snippet

const calculateSavingsRate = (monthly_income, monthly_fixed_expenses) => {
  return ((monthly_income - monthly_fixed_expenses) / monthly_income) * 100;
};

const calculateEmergencyFundCoverage = (current_savings, monthly_fixed_expenses) => {
  return current_savings / monthly_fixed_expenses;
};

const calculateFinancialIndependenceNumber = (monthly_fixed_expenses, withdrawalRate = 0.04) => {
  return (monthly_fixed_expenses * 12) / withdrawalRate;
};

const calculateExpenseToIncomeRatio = (monthly_income, monthly_fixed_expenses) => {
  return (monthly_fixed_expenses / monthly_income) * 100;
};

// Wealth-building metrics
const calculateNetWorth = (current_savings, invested_amount) => {
  return current_savings + invested_amount;
};

const calculateWealthRatio = (netWorth, annual_income) => {
  return (netWorth / annual_income) * 100;
};

const calculateInvestmentRatio = (invested_amount, netWorth) => {
  return (invested_amount / netWorth) * 100;
};

// Financial health scoring
const calculateFinancialHealthScore = (metrics) => {
  let score = 0;

  // Emergency fund score (0-25 points)
  const emergencyFundScore = Math.min(metrics.emergencyFundCoverage * 4.17, 25);

  // Savings rate score (0-25 points)
  const savingsRateScore = Math.min(metrics.savingsRate, 25);

  // Investment ratio score (0-25 points)
  const investmentRatioScore = Math.min(metrics.investmentRatio * 0.5, 25);

  // Expense to income ratio score (0-25 points)
  const expenseRatioScore = Math.max(0, 25 - (metrics.expenseToIncomeRatio * 0.5));

  score = emergencyFundScore + savingsRateScore + investmentRatioScore + expenseRatioScore;

  return {
    total: score.toFixed(2),
    breakdown: {
      emergencyFund: emergencyFundScore.toFixed(2),
      savingsRate: savingsRateScore.toFixed(2),
      investmentRatio: investmentRatioScore.toFixed(2),
      expenseRatio: expenseRatioScore.toFixed(2)
    }
  };
};

3. Personalized Insights

We go beyond numbers by interpreting your data and offering tailored recommendations. Our insights guide you to:

Maximize your savings potential.
Optimize your investment strategies.
Build a solid financial foundation.

4. Visualize Your Finances

Our system converts your data into intuitive visuals:

Investment Growth Graphs: Understand the growth of your investments over time through lumpsum or SIP strategies.
Spending Trends: Identify patterns and areas for improvement in your expenses.

Visual graph creation API (code snippet)

def calculate_investments(
    lump: float = Query(..., ge=0, description="Lump sum investment amount (must be greater than or equal to 0)"),
    monthly: float = Query(..., ge=0, description="Monthly SIP investment amount (must be greater than or equal to 0)")
):
    if lump == 0 and monthly == 0:
        raise HTTPException(status_code=400, detail="At least one of lump sum or monthly investment must be greater than 0")

    time_years = np.arange(1, 26)
    rates = [0.10, 0.12, 0.15]
    colors = {"10%": "blue", "12%": "yellow", "15%": "red"}

    # Calculate lump sum growth
    lump_data = {"Year": [], "Amount": [], "Rate": []}
    if lump > 0:
        for rate in rates:
            for year in time_years:
                amount = compound_interest(lump, rate, year)
                lump_data["Year"].append(year)
                lump_data["Amount"].append(amount)
                lump_data["Rate"].append(f"{int(rate * 100)}%")
        lump_df = pd.DataFrame(lump_data)
        lump_graph = generate_graph(lump_df, "Lump Sum Investment Growth Over 25 Years", colors)
    else:
        lump_graph = None

5. Financial Podcasts Tailored to You

To make your financial journey even more engaging, we generate a personalized podcast based on your financial report. These podcasts include actionable advice inspired by renowned financial books like Rich Dad Poor Dad and The Intelligent Investor. This feature makes it easier to understand your report and implement the suggestions effectively.

6. Generate and Deliver Reports

Finally, all insights, visuals, and recommendations are compiled into a clean, easy-to-read report. Delivered straight to your inbox, it’s designed to help you make confident financial decisions without the need for spreadsheets or jargon.

With FreedomFund, financial literacy becomes accessible, actionable, and even enjoyable. From personalized insights to podcasts inspired by the greatest financial minds, we’re here to guide you toward financial independence and clarity.

Architecture

Demo

Freedomfund: agent.ai
Visit our website to learn more and get started on your journey to financial freedom!

Website Screenshots

Agentai WorkFlow

Result Example

Code

Ritesh2351235 / freedomFund

FreedomFund is a personalized financial literacy platform that generates detailed reports, offering insights and recommendations to improve financial health. It uses AI tools to analyze data, visualize key metrics, and simplify finance management. The platform also provides podcasts based on popular financial books to enhance understanding.

FreedomFund - Your Personal Financial Literacy Agent

Overview

FreedomFund is a personalized financial literacy agent designed to help individuals understand and manage their finances better. Our platform generates detailed financial reports, provides insights, and offers recommendations to optimize savings, investments, and overall financial health. With the power of financial literacy, we aim to guide you toward achieving financial independence.

Features

Personalized Financial Reports: Get customized reports based on your financial data, including savings rate, emergency fund coverage, and investment analysis.
Actionable Insights: Receive clear recommendations and insights based on your financial health.
Visualizations: Easily understand your financial status through graphs and charts generated from your data.
Podcast: Access a podcast based on the financial reports, along with suggestions from famous financial books like Rich Dad Poor Dad and The Intelligent Investor to help you understand your report more clearly.
Real-time Updates: Get accurate and timely…

View on GitHub

Agent.ai Experience

Building FreedomFund on AgentAI has been an exciting experience. The platform’s powerful tools and features made it easy to bring our idea of personalized financial literacy to life. From handling workflows to integrating AI-driven insights, AgentAI simplified much of the process, allowing us to focus on creating value for users.

However, we did face some challenges, like integrating webhooks and formatting HTML for reports. While the tools were helpful, more detailed documentation and examples would have made things smoother.

Overall, AgentAI impressed us with its capabilities and potential. With a few improvements in developer resources, it can become an even better platform. We’re excited to keep building and improving FreedomFund with AgentAI!

This project was built in collaboration with my friend, @gocool_777. A big thanks to him for his valuable contributions and support!

FluentAI: Your All-in-One Solution for Mastering English Communication ✨

Ritesh Hiremath — Sun, 24 Nov 2024 18:32:35 +0000

This is a submission for the AssemblyAI Challenge : Sophisticated Speech-to-Text.

What I Built

FluentAI is a cutting-edge AI-powered application designed to make English learning fun, engaging, and personalized. Whether you want to refine your reading, writing, or speaking skills, FluentAI offers tailored exercises, instant grading, and detailed feedback powered by advanced AI. It’s your one-stop solution to transform how you communicate in English!

How it works

Sign Up and Choose Your Focus

Start by signing up and creating your profile. Your details are securely stored in a PostgreSQL database for tracking your progress and storing feedback. Once logged in, select the area you want to work on: Reading, Writing, or Speaking.

Reading Practice

Read aloud the given passage and record your audio.
Once submitted, the audio file is processed:
- Transcribed into text using AssemblyAI Universal.
- Sent to Ollama Mistral Model for analysis.
Receive scores and feedback on parameters like grammar, pronunciation, and tone of speech, all of which are updated in the database for future tracking.

Speaking Practice

Choose a topic (or freestyle about your day!).
Record and submit your audio.
The same transcription and analysis process is applied to assess your fluency, clarity, and intonation.
Personalized feedback and grades are stored in the database to help you chart your improvement.

Writing Practice

Submit your written piece—be it an essay, story, or just a random thought.
Your submission is sent to Ollama Mistral Model for a detailed evaluation based on parameters like structure, grammar, and coherence.
Feedback is updated in the database to help you keep track of your writing progress.

PS: All your scores and feedback live safely in the database—no public displays of your grammar flubs or poetic masterpieces. 😅

Demo

checkout the live demo here:

fluentAi

Disclaimer: The responses may take around 5-10 min to generate as the server is deployed on AWS t2.large
(PS : I'm broke hehe..)

Landing Page

Reading section

Writing section

Speaking section

Feedback section

Source Code:

Ritesh2351235 / Fluentai

FluentAI is an AI-driven platform to enhance English communication skills through personalized practice in reading, writing, and speaking. It provides detailed feedback and progress tracking to help users improve effectively.

FluentAI

FluentAI is an AI-powered application that helps users improve their English communication skills by practicing reading, writing, and speaking. The app provides personalized feedback and grades for each exercise, helping users identify areas of improvement and track progress over time.

Features

Reading Practice: Read passages aloud, get feedback on grammar, pronunciation, and tone.
Speaking Practice: Speak on a topic, receive scores on fluency, clarity, and intonation.
Writing Practice: Submit written content and get detailed feedback on structure, grammar, and coherence.

Tech Stack

Frontend:
- React
- Tailwind CSS
- Shadcn Components
- GSAP (for animations)
Backend:
- Node.js
- Express
- AssemblyAI API (audio transcription)
- Ollama Mistral Model (AI evaluation and feedback)
Database & Hosting:
- PostgreSQL (Aiven) for user data and progress tracking
- AWS EC2 (t2.large) for backend deployment
- Vercel for frontend deployment

Feel free to contribute or report issues to help enhance FluentAI!

View on GitHub

Journey

FluentAI was built to address a pressing challenge: providing an affordable, interactive way for users to improve their English communication skills. The integration of AssemblyAI’s Universal Speech-to-Text Model was crucial in bringing this vision to life.

AssemblyAI is the backbone of the audio transcription process, ensuring high accuracy and quick results.
The Ollama Mistral Model acts as the smart assistant, analyzing text and providing context-aware feedback.
It wasn’t just about building an app—it was about creating a seamless experience where users could learn, grow, and gain confidence in their abilities.

Tools Used

Building FluentAI involved a stack of powerful tools and technologies:

Frontend

React: For crafting a dynamic and responsive user interface.
Tailwind CSS: A utility-first framework to ensure sleek and modern designs.
Shadcn: For reusable and aesthetic UI components.
GSAP: For animations that make the app lively and engaging.

Backend

Express + Node.js: For handling API requests and business logic.
AssemblyAI: Converts audio into text with remarkable accuracy.
Ollama Model (Mistral): Analyzes transcripts and text for personalized grading and feedback.

Infrastructure

AWS EC2 (t2.large): For robust and scalable server hosting.(I'm Kinda broke, but this does the job..)
Aiven Database: Stores user-profiles and progress securely.
Vercel: Deploys the frontend for a fast and reliable user experience.

Final Thoughts

Building FluentAI has been an incredible journey, blending AI, creativity, and technology to address a real-world need. Whether you’re prepping for a presentation, working on essays, or simply polishing your conversational skills, FluentAI has your back.

With AI-driven grading and personalized feedback, learning English is no longer daunting—it’s a journey of constant growth and discovery.

Future Plans

I aim to make FluentAI even smarter with features like:

Daily progress reports.
AI-curated exercises based on your learning style.
Gamified challenges to keep the learning fun and rewarding.

NutritionAI: Your Personal AI-Powered Nutrition Guide for Smarter Eating!

Ritesh Hiremath — Sun, 10 Nov 2024 12:09:23 +0000

This is a submission for the Open Source AI Challenge with pgai and Ollama

What I Built

NutritionAI is an AI-powered application that helps users track their dietary habits by allowing them to upload images of their meals. The app analyzes these images to provide detailed nutritional insights and personalized recommendations based on individual dietary goals and preferences.

How it Works

Diet Preferences: Users first input their dietary goals and health needs, which are sent to the backend. This data is embedded using the Mistral model and stored in a pgvector database.

This is the schema for the user's diet preferences

CREATE TABLE users (
    id SERIAL PRIMARY KEY,
    age INT NOT NULL,
    height FLOAT NOT NULL,
    weight FLOAT NOT NULL,
    caloric_target INT NOT NULL,
    protein_target INT NOT NULL,
    dietary_preferences TEXT[],  
    complications TEXT[],      
    embedding VECTOR             
);

Meal Upload & Image Analysis: Once on the main page, users can upload a meal image. The backend processes the image and sends it to the Llava model, which identifies food items and dishes.
Nutritional Analysis: The identified food items are passed to the Mistral model to generate nutritional information. These stats are embedded as vectors, saved in the database for future use.

This schema represents the food items.

CREATE TABLE food_items (
    id SERIAL PRIMARY KEY,
    name VARCHAR(255) NOT NULL,   
    embedding VECTOR,              
    nutrition_info JSONB          
);

Similarity Check: The new meal vector is compared with the user's dietary preferences using cosine similarity (via pgai) to evaluate alignment with the user's goals.
Personalized Recommendations: Based on similarity scores, NutritionAI analyzes the nutritional data and suggests diet adjustments that best meet the user's preferences and health objectives.

Architecture

Demo

Here's the link to the working project
nutrition-ai-bay

Disclaimer: The generation of nutritional stats and personalized recommendations may take around 5-10 minutes, as the Node.js application is hosted on an Amazon EC2 t2.large instance.

Source code:

Ritesh2351235 / nutrition-ai

NutritionAI is an AI-driven application that allows users to upload meal images for nutritional analysis. It provides personalized dietary recommendations based on user profiles and food item embeddings.

NutritionAI

This is a submission for the Open Source AI Challenge with pgai and Ollama

NutritionAI is an innovative application designed to help users achieve their dietary goals by providing personalized nutritional insights. By allowing users to upload pictures of their daily meals, the app analyzes the nutritional content and offers tailored recommendations based on individual dietary needs.

NutritionAI simplifies the journey to healthier eating by enabling users to sign up easily, upload meal images, and receive detailed nutritional statistics that align with their diet goals.

Key Features

Sign Up Form: Create a personalized account to track your dietary journey.
Meal Upload: Easily upload pictures of your meals for analysis.
Nutritional Insights: Receive detailed nutritional information for each meal.
Personalized Recommendations: Get tailored dietary suggestions based on your profile and preferences.
Daily Tracking: Track your nutritional intake and progress towards goals.
Health Considerations: Recommendations…

View on GitHub

Screenshots

Diet Preferences Page

Main Food Analyzer Page

Tools Used

React: Framework for building dynamic user interfaces.
Tailwind CSS: Utility-first CSS framework for styling.
Node.js: JavaScript runtime for backend development.
Express: Framework for handling API requests and routing.
Multer: Middleware for processing file uploads.
Ollama Models:
- Llava Model: Analyzes meal images to identify dishes.
- Mistral Model: Generates nutritional content and embeddings.
TimescaleDB (PostgreSQL): Database for storing user and nutritional data.
pgvector: Utilized for storing and querying vector embeddings of user profiles and food items, enhancing the accuracy of recommendations.
pgai: Integrated to leverage AI capabilities for generating insightful nutritional content and recommendations based on user data.
Amazon EC2 (Ubuntu t2.large): Hosts the Node.js application.
Vercel: Deploys the frontend application efficiently.

Final Thoughts

Building NutritionAI has been an exciting project(PS: Took me 5 whole days to build this), especially with the challenges of implementing real-time image analysis and generating personalized nutritional recommendations through advanced AI models. With Ollama’s models and PostgreSQL extensions, I was able to create an engaging, user-friendly experience that brings detailed nutritional insights to users' fingertips.

In the future, I plan to enhance NutritionAI with additional features, such as a personalized fitness plan that aligns with users' dietary habits, and tailored meal planning suggestions. Improvements could also include faster processing times and even more advanced dietary analytics.

Prize categories:
Open Souce Models from Ollama
All the extensions.

Hacktoberfest 2024: My Open Source Journey.

Ritesh Hiremath — Sun, 03 Nov 2024 17:06:06 +0000

This is a submission for the 2024 Hacktoberfest Writing challenge: Contributor Experience

Just a month ago, I didn’t know much about GitHub other than the basics of storing my projects and the classic three commands to push changes. Open source sounded exciting but overwhelming to me.
I had no idea how to start or what "good" contributions looked like. Then came Hacktoberfest 2024, and with it, my first PR.

As a first-time participant, Hacktoberfest became a whirlwind of learning experiences. I worked on contributions with organizations like FreeCodeCamp, JSON Schema, React-Chatbotify, and Layer5. Each one taught me something new: version control, communicating with open-source maintainers, navigating Slack communities, and the essentials of a good PR. I even began to appreciate the significance of open source beyond just “coding” – it’s a collaborative, evolving effort to make software more accessible and robust.

October turned out to be incredibly productive and memorable. Hacktoberfest gave me the kickstart I’d been hesitant to take, and it couldn’t have come at a better time. As I wrapped up my PRs, I also celebrated my birthday, feeling a great mix of pride and excitement for all that I’d accomplished.

This experience wasn’t just about ticking off a few contributions; it marked the beginning of a path I feel genuinely enthusiastic about. Here's to many more PRs, learning, and community connections!
Thank you HacktoberFest 2024!!!!

Here are some insights!!!