Aspiring Cybersecurity Expert | TryHackMe Enthusiast | C++ & Python Developer | AI Prompt Engineer | Student Ambassador

medium

youtube

github

KUNJ SHAH • Peerlist

Every product needs a walkthrough video, and making one takes hours. You write a script, record the screen, retake it, record the voice-over and sync the captions. Then you ship one update and the video is already out of date.Walkcast does it from a single URL. It opens a real Chromium browser, signs in if you give it credentials, walks through your pages and records what actually happens. Every line of narration is written from recorded events, not from a script and not guessed by an AI. You get:<ul><li>recording.mp4 with a spoken voice-over (H.264 + AAC)</li><li>captions.vtt timed to every line</li><li>A clickable transcript with per-page chapters</li><li>narration.wav, plus a full action trace</li></ul>You pick the voice: the free built-in one, ElevenLabs, OpenAI, Google, Azure, Polly, Sarvam, and more, or your own microphone. Each voice can be sampled before you record.Walkcast Lab uses the same recorder for a second job: debugging browser AI agents. It turns a failed agent run into a reproducible regression test. The cause is classified from the application’s real state, not from the agent’s claim or an LLM judge, and anything the rules can’t explain is marked “unexplained”. Replay that test on the next agent version to see whether it fixed the bug or broke something else.Built-in guardrails:<ul><li>You must confirm you own the site or are authorized to test it.</li><li>It never bypasses CAPTCHA or 2FA. It stops and says so.</li><li>It honours robots.txt.</li><li>Passwords are never written to disk or spoken in the narration.</li></ul>Open source under the MIT license, and it costs nothing. Available for Windows, macOS and Linux.Categories / tags Developer Tools, Productivity, Open Source, AI, Video, Testing, MarketingLinks<ul><li>Website: <a target="_blank" rel="noopener noreferrer nofollow ugc" class="break-all link-break text-gray-500 hover:underline" href="https://walkcast.kunjshah.in">https://walkcast.kunjshah.in</a></li><li>GitHub: <a target="_blank" rel="noopener noreferrer nofollow ugc" class="break-all link-break text-gray-500 hover:underline" href="https://github.com/KunjShah95/walkcast">https://github.com/KunjShah95/walkcast</a></li></ul>

GitHub - KunjShah95/walkcast · GitHub

A curated index of the infrastructure behind working AI systems — 112 tools across 10 sections, 112 per-tool pages, 9 comparisons, 5 symptom checklists, 51 glossary terms and 11 essays on the architectural decisions behind them.Sections are ordered as a production stack: layer 1 is the substrate everything else runs on, layer 9 is the surface you look at. Off-stack material (reading, courses) sits deliberately outside the stack.

lattice

<h3>AI-native workspace for engineers.</h3><ul><li>Notes: Markdown with frontmatter, tags, project assignment, pinning, and [[wikilinks]] that resolve to backlinks.</li><li>Tasks: Kanban board (Backlog / Todo / In Progress / Done), priority, due date, and linked notes.</li><li>Projects: Notes, tasks, timeline, and resources under one roof with visible progress.</li><li>Daily notes: Auto-created daily notes with sections for goals, journal, tasks, learnings, wins, problems, and tomorrow. Unfinished tasks roll over automatically.</li><li>Resources: Store and query code snippets, bookmarks, reading lists, architecture notes, and meeting notes.</li><li>Quick Capture: Use a command palette (⌘K) or Quick Capture button to instantly save notes or tasks. Auto-triage rules can route captures by keyword.</li><li>Search: Instant search across notes, tasks, and projects via the command palette. Semantic search allows for meaning-based retrieval, falling back to local keyword scoring without an API key.</li><li>AI Assistant: Ask questions about your workspace in plain English, with answers citing the exact notes they came from. Threads are persisted for continued conversations.</li><li>Knowledge Graph: Visualize how notes connect via wikilinks, task links, and project membership with a force-directed graph.</li><li>PDF Chat: Upload documents and ask questions about them, receiving cited answers.</li><li>Voice Notes: Record and transcribe voice notes, with local fallback transcription.</li><li>Mind Maps: Convert any note into a visual mind map.</li><li>Automation: Features include recurring tasks, auto-triage for captures, and daily rollover of unfinished tasks. Background work is handled by a client-drained queue.</li><li>Integrations: Connect with GitHub to import issues into tasks and link repositories to projects. Calendar export for tasks is also available.</li></ul>The stack includes Next.js 16 (App Router), React 19, TypeScript, Supabase (Postgres + pgvector + RLS), Supabase Auth, TanStack Query, Zustand, shadcn/ui, Tailwind CSS v4, and Framer Motion. AI features utilize OpenAI embeddings and chat, with optional local fallbacks.This project is open source and designed to be a unified system for engineers to manage their knowledge, tasks, and projects, reducing context switching and improving information retrieval.

EngineerOS

Most portfolios show projects. This one showcases systems.I built this portfolio as a central hub for AI engineering work, focusing on:<ul><li>Agent-based architectures</li><li>Real-world problem solving</li><li>System design thinking</li></ul>Instead of static project cards, the goal was to create a platform that: → Demonstrates how systems work → Highlights technical depth → Converts visitors into opportunities

Kunjs Portfolio Website

AI is already making healthcare decisions.But most models aren’t tested for bias.That’s dangerous.I built EquityLens — a system that audits AI for fairness, compliance, and explainability.This isn’t optional anymore.Breakdown 👇 github.com/KunjShah95/EquityLens

EquityLens

Transform Developer Onboarding with AI-Powered Learning Paths, Real-Time Analytics, and Accelerated First Contributions<h3>What is CodeFlow?</h3>CodeFlow is an enterprise-grade AI-driven platform that significantly reduces developer onboarding time by 50-70% through:<ul><li>Real-time Team Analytics - Identify where developers face challenges</li><li>AI-Powered Learning Paths - Tailored to each developer's skills</li><li>Knowledge Verification - Adaptive quizzes with spaced repetition</li><li>First PR Acceleration - Guide developers to their first contribution swiftly</li><li>Living Playbooks - Capture institutional knowledge from senior developers</li><li>Smart Coaching - Real-time guidance and gamification to motivate juniors</li></ul>Result: Junior developers reach productivity 75% faster | $37.5K savings per developer | 15:1 ROI<h3>Key Features</h3><ul><li>Team Onboarding Dashboard - Track onboarding metrics in real-time</li><li>Knowledge Verification Quizzes - AI-generated quizzes from learning modules</li><li>First PR Acceleration Mode - Finds beginner-friendly GitHub issues</li><li>Collaborative Playbooks - Senior developers create "How to..." guides</li><li>Progress Coach with Gamification - Daily guidance and progress tracking</li><li>Advanced Analytics Engine - Architecture visualization and efficiency benchmarking</li></ul>

GitHub - KunjShah95/AHS-2026

Explore the capabilities of kunjs-PORTFOLIO-WEBSITE, a dynamic portfolio platform hosted on Vercel. This project is designed to showcase personal achievements and projects with a modern and responsive design. Built using TypeScript, JavaScript, CSS, and HTML, it leverages the power of Vite for fast development and Tailwind CSS for styling. The project is open source, allowing developers to contribute and enhance its features. Key features include a customizable layout, integration with GitHub for seamless updates, and a user-friendly interface. Whether you're a developer looking to showcase your work or someone interested in building a personal website, this project offers a robust foundation.

GitHub - KunjShah95/kunjs-PORTFOLIO-WEBSITE

GapMiner is an innovative AI-powered research tool designed to assist researchers in discovering unsolved research problems from academic papers. By leveraging advanced technologies such as Google Gemini for intelligent analysis and Firecrawl for web scraping, GapMiner automatically extracts limitations and identifies research gaps, offering researchers new opportunities for exploration. The platform supports intelligent web scraping from sources like arXiv, OpenReview, and ACL Anthology, and categorizes gaps into data, compute, evaluation, and methodology types. With secure authentication via Firebase Auth and cloud storage using Firestore, users can manage their research findings efficiently. The modern, responsive interface, built with React and styled with Tailwind CSS, ensures a seamless user experience, complete with dark mode and smooth animations. GapMiner is an essential tool for researchers aiming to push the boundaries of academic inquiry.

GitHub - KunjShah95/GAPMINER

A sleek, full-stack platform designed to boost job-seekers’ chances by integrating resume parsing, ATS scoring, job-matching and cover-letter generation — all powered by modern LLMs and vector-search under the hood.What it does<ul><li>Parses resumes (PDF/DOCX/TXT) including OCR for scanned documents.</li><li>Scores resumes using an ATS-style 100-point system with actionable feedback.</li><li>Matches a candidate’s resume to job descriptions, providing compatibility insights.</li><li>Generates personalized cover letters and resumes from scratch or via templates.</li><li>Offers job-search integration, skill-gap analysis and other productivity modules.</li><li>Built as a multi-page UI (Streamlit &amp; Gradio), deployable via Docker, cloud or local.</li></ul>Why it's useful In today’s competitive job market, standing out depends not just on experience but on how your materials are presented and aligned with each role. job-snipper helps automate and elevate that process: offering data-driven feedback, AI-enhanced matching and document generation — saving time and improving match quality.Tech stack at a glance<ul><li>Front-end: Streamlit 1.50+, Python 3.13+</li><li>Models &amp; AI: Google Gemini, Groq, OpenAI, Anthropic via LangChain</li><li>Data-processing: pandas, numpy; Visualization: seaborn, matplotlib, plotly</li><li>Document handling: PyPDF2, python-docx, pdfplumber; OCR via Mistral &amp; Tesseract</li><li>Storage &amp; Deployment: ChromaDB vector store; Docker / Docker Compose; Cloud friendly</li><li>Best practices: Local environment variables for keys, robust security and testing in place.</li></ul>Getting started<ol><li>Clone this repo: git clone <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="https://github.com/KunjShah95/job-snipper.git">https://github.com/KunjShah95/job-snipper.git</a></li><li>Create a virtual environment (e.g., python3 -m venv .venv → source .venv/bin/activate)</li><li>Install dependencies: pip install -r requirements.txt</li><li>Copy .env.example → .env, add your API keys (Groq recommended, or Gemini, OpenAI)</li><li>Run the app: streamlit run <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="http://Home.py">Home.py</a> and open <a target="_blank" rel="noopener noreferrer" class="break-all link-break text-gray-500 hover:underline" href="http://localhost:8501">http://localhost:8501</a></li></ol>Who should use it / contribute<ul><li>Job-seekers looking for smarter ways to craft or refine their resumes and cover letters.</li><li>Career coaches, HR professionals, or recruiters needing tools to analyse and match applicant documents.</li><li>Developers or open-source contributors who want to extend AI workflows, add new modules (e.g., multi-language support), or improve the matching and scoring algorithms.</li></ul>Future vision Planned enhancements include multi-language support, mobile native apps, integration with LinkedIn and major job boards, enhanced career-path recommendations, video resumes and more.

GitHub - KunjShah95/job-snipper

NEXUS AI Terminal Assistant is a professional, multi-model AI terminal application designed to enhance productivity and research through seamless, secure, and powerful natural language AI capabilities. This innovative tool supports a variety of AI models including Google Gemini, Groq, Ollama, HuggingFace, and DeepSeek, allowing users to effortlessly switch between them for optimal performance. Key features include comprehensive input validation, API key encryption, and command safety, ensuring robust security. The application also offers state-of-the-art speech-to-text and voice command support, integrated real-time web search, and persistent conversation memory with ChromaDB. Its modern terminal interface, powered by Rich, provides a beautiful UI with status indicators and helpful banners. NEXUS is cross-platform, running on Windows, Mac, and Linux, with Docker support included. As an open-source project under the MIT license, it offers a modular codebase ready for contributions, making it an extensible solution for developers and researchers alike.

GitHub - KunjShah95/NEXUS-AI.io

<h3>Overview</h3>Student Benefits Hub is a cutting-edge, interactive 3D web application designed to showcase over 150 premium student benefits valued at more than $200,000. Built with modern web technologies, this application offers a performance-optimized experience with advanced CSS3 animations and a responsive design that works seamlessly across all devices. The platform aggregates a wide range of student benefits, discounts, and free services, presented in an engaging 3D environment.<h3>Key Features</h3><ul><li>Modern ES6+ JavaScript: Provides modular and maintainable code with async/await patterns.</li><li>CSS3 Advanced Animations: Utilizes hardware-accelerated transforms and transitions for smooth animations.</li><li>3D Rendering Engine: Features a custom-built CSS3 3D transform system for an immersive experience.</li><li>Component-Based Design: Offers reusable UI components with consistent styling.</li><li>Progressive Enhancement: Ensures graceful degradation for older browsers.</li></ul><h3>UI/UX Design System</h3><ul><li>Material Design 3.0: Adheres to Google's latest design guidelines.</li><li>Glass Morphism: Incorporates modern backdrop-filter effects with transparency layers.</li><li>Micro-interactions: Includes carefully crafted hover states and click animations.</li><li>Typography Scale: Utilizes systematic font sizing using a modular scale.</li><li>Color Theory: Employs a scientifically chosen color palette for optimal readability.</li></ul><h3>Performance Optimizations</h3><ul><li>Critical Path Optimization: Prioritizes above-the-fold content for faster rendering.</li><li>Lazy Loading: Uses the Intersection Observer API for efficient resource loading.</li><li>Animation Performance: Achieves 60fps animations using transform and opacity only.</li><li>Memory Management: Ensures efficient event listener cleanup and DOM manipulation.</li><li>Bundle Size: Implements a zero-dependencies, pure vanilla implementation.</li></ul>

GitHub - KunjShah95/StudentHub.com

Title: Train Booking Management System (C++)

Description:
This Train Booking Management System is a comprehensive, menu-driven console application built in C++ for managing train bookings, customer information, ticketing, and payments. It provides a seamless user experience by allowing operations like adding and editing train schedules, customer management, booking and modifying tickets, and processing payments.

Key features of the system:

Train Management: Add, edit, and display train details including schedules, types, and routes.
Customer Management: Manage customer profiles with loyalty points and contact details.
Booking System: Create, modify, and cancel bookings, linked to specific customers and trains.
Ticketing System: Issue and print tickets with seat assignments and pricing.
Payment Processing: Handle payments and refunds for booking transactions.
Route Management: Define and manage train routes with multiple stops.
The system uses file handling for persistent data storage and supports multiple operations within a structured, easy-to-use menu. This project demonstrates proficiency in object-oriented programming, file handling, and interactive C++ applications.

Technologies Used:

C++
File Handling
Object-Oriented Programming (OOP)