Forem: Arslan Sarwar

KaneAI: AI Agent for E2E Software Testing

Arslan Sarwar — Fri, 29 Nov 2024 19:51:12 +0000

The constant drive towards automation pushes Generative AI (GenAI) into the evolving software development and testing world. GenAI can help software testers with many tasks throughout the Software Testing Life Cycle (STLC). The primary purpose of the STLC is to identify any defects in the software application as early as possible.

GenAI is revolutionizing the approach towards testing by allowing AI from test planning to test execution. This revolution marks a radical enhancement in the testing scope which significantly reduces the need for human intervention.

Here's what you're going to explore in this guide:

Introducing KaneAI: AI-powered End-to-End Automation Testing Platform
Testing Workflow of KaneAI
Key Features of KaneAI
Conclusion
Additional Resources

Introducing KaneAI: AI-powered End-to-End Automation Testing Platform

KaneAI is an advanced AI-powered platform built with a unique approach to create, debug, and evolve end-to-end tests using natural language. It enables testers to run automated tests on web, mobile, and cloud-based applications using prompts.

One of the main goals of KaneAI is to democratize quality assurance by making it easier for people with different skill sets to participate in the testing process.

KaneAI leverages natural language processing (NLP) and GenAI technologies to understand human instructions and generate test scripts. This approach differs from traditional keyword-driven automation or BDD frameworks that still require a significant learning curve and technical expertise.

KaneAI supports popular test automation frameworks such as Selenium, Appium, and Playwright. This allows users to export generated tests in their preferred programming language and integrate them with existing codebases.

LambdaTest has extensive experience in test execution infrastructure and provides insights from running tests at scale. KaneAI is tightly integrated with LambdaTest's HyperExecute platform, which offers features such as parallel testing, load balancing, and fail-first execution.

KaneAI prioritizes visual testing to reduce test flakiness. It verifies the presence and visibility of elements in the DOM before interacting with them. This approach simulates end-user behavior and helps identify issues that locator-based testing might miss.

KaneAI offers AI-powered debugging features such as error classification, flaky test analysis, and root cause analysis. These features help teams quickly triage and fix test failures by identifying patterns and providing insights into potential problem areas.

KaneAI provides test intelligence and observability features. It generates reports, dashboards, and insights from test execution data to give a comprehensive overview of application health and identify trends that can help predict future failures.

Testing Workflow of KaneAI

Understand human instructions to generate test cases.
Extract key requirements from Jira tickets, including user outcomes, user instructions, and target application URLs.
Test applications behind authentication and handle dynamic parameters.
Seamlessly integrate with LambdaTest’s test planner.
Intelligently analyze each test step, ensuring alignment with the test environment.
Provide real-time feedback and adapt to potential issues that might arise.
identify bugs and provide a summary view of the testing workflow.
Users can edit test cases in KaneAI's intuitive "playground mode."
Help users save different versions of test cases.
Allow test cases to be exported in multiple programming languages and across different test automation frameworks.
Deeply integrate with HyperExecute and other CI/CD platforms.
Support auto-healing of test cases through runtime code patching.
Enable continuous testing through single-click execution.
Convert actions into natural language instructions to generate bullet-proof tests.
Express sophisticated conditions and assertions in natural language.
Run generated tests across 3000+ combinations of browsers, OS, and real devices.
Detailed test execution reports with Test Intelligence and deep analytics.

Key Features of KaneAI

Universal Language: Instruct KaneAI using natural language, eliminating the need for complex scripting.
Speed and Efficiency: Automate testing processes, freeing up time for innovation and creativity.
Self-Adaptation: Adapt to changes in the application, ensuring tests stay in sync with development.
Bug Detection: Uncover bugs that manual testing might miss.
Versatility: Support testing across various environments, including web and mobile.
Simplicity: Simplify complex testing scenarios, making testing more accessible.

Conclusion

KaneAI is the world's first end-to-end software testing agent. It is a game-changer tool in the software testing world. It frees your time for creative and innovative work instead of dealing with repetitive testing tasks. It gives you a competitive edge in the world of speed and quality.

KaneAI supports complex workflows through built-in features including intelligent test planning, AI-powered test healing, and advanced conditional assertions. It helps you ship better software faster With AI-powered insights and seamless integration with modern workflows.

KaneAI provides an efficient and flexible testing process as you can run tests sequentially or with multiple prompts. It integrates seamlessly with existing workflows, supporting popular tools such as Jira, Slack, GitHub Actions, and Microsoft Teams.

KaneAI is currently in early beta and access is being rolled out progressively. Interested users can sign up for the beta program on the LambdaTest website!

Additional Resources

Full-Stack Starter Kit: Astro + Tailwind CSS + Neon Postgres + Drizzle ORM + TypeScript + React(Optional)

Arslan Sarwar — Mon, 02 Sep 2024 05:07:39 +0000

This is a submission for the Neon Open Source Starter Kit Challenge : Ultimate Starter Kit

My Kit

This Starter Kit includes all the basics you need to build a full-stack application using Astro and Neon Postgres Database. You can follow the Step-by-Step guide in the following Starter Kit for building Full-Stack applications using Astro + Neon Postgress Database.

Link to Kit

Full-Stack Starter Kit

Your Journey

Astro is amazing. You can start simple and add complexity as the project grows. You can build simple (SSG) websites to complex full-stack apps using Astro. With the addition of stable astro actions in the framework, You can achieve more with this framework. So, here's the thing:

"Astro powers the world's fastest marketing sites, blogs, e-commerce websites, and more." Source

Neon is amazing as well just like Astro.

"Ship faster with Postgres.
The database you love, on a serverless platform designed to help you build reliable and scalable applications faster." Source

Here's why I've chosen the following tech stack for this Starter Kit Challenge by Neon:

Astro: Start Simple and add complexity as the project grows.
Neon: No need to worry about the scalability of the Postgres Database.
Drizzle ORM: Lightweight and performant TypeScript ORM with the developer experience in mind.
Tailwind CSS: Quickly build your desired UI with the Tailwind classes.
React(Optional): Integrate React Components with the Astro Components if you want to.

Wordle Game Built using FARM Stack - MongoDB Atlas Hackathon 2022 on DEV

Arslan Sarwar — Tue, 06 Dec 2022 21:06:43 +0000

What I built

Popular Wordle Game using FARM (FastApi, React, MongoDB) Stack

Category Submission:

(Think Outside the JS Box)

App Link

Wordle Game

Screenshots

Description

You have to guess 5 letters word in maximum 6 turns.

How to Play:

Submit word by pressing Enter after adding 5 letters.
Press Backspace to remove the letter.
You can't repeat the words during multiple guesses.
You have to submit exact 5-letter word.

Word Hints:

Tile gets green if letter is at the right place.
Tile gets yellow if letter is in the word but not at right place.
Tile gets gray if letter is not in the word.
All the used letters are printed on the keypad with colors (green, yellow, gray) as well.

Link to Source Code

Source Code

Permissive License

(MIT License)

Background

(What made you decide to build this particular app? What inspired you?)
Being non-native english speaker, I love this game. It challenges me for my english vocabulary. I play this game a lot. I always work in JS Stack. But this hackathon motivated me to try another stack outside JS. So, I built this game using FARM stack.

How I built it

(How did you utilize MongoDB Atlas? Did you learn something new along the way? Pick up a new skill?)
I stored all the words for my game in the MongoDB Atlas Database. I learned how to use MongoDB with Python. I learned FastApi and Python during this hackathon.

Frontend --> React
Backend --> FastApi
Database --> MongoDB Atlas
Deployment --> Vercel

Additional Resources/Info

Game Demo

Article on FARM Stack by MongoDB

Youtube Video on FARM Stack by MongoDB

MongoDB Integration with Vercel by MongoDB