mongodb
AI Research Workspace
Most of my earlier projects were built around Express.js for the backend. While working on AI-focused applications, I realized that building serious AI systems requires a different backend ecosystem and a deeper understanding of how AI workflows actually operate. That led me to learn FastAPI, and I used this project to not only learn it, but also polish my backend and AI engineering skills by building something from the ground up. Introducing 🚀 AI Research Workspace (AIRW) — an AI-powered workspace designed to interact with and retrieve information from different types of user data. 🔹 What I explored and implemented: • FastAPI for the backend and API architecture • LangChain & LangGraph for building structured AI workflows and agentic execution • RAG (Retrieval-Augmented Generation) for document-based question answering • Document chunking, embeddings, vector search & hybrid retrieval • Semantic retrieval using embeddings combined with traditional keyword-based retrieval • Query rewriting and retrieval pipelines to improve search quality • MongoDB Atlas Vector Search for storing and retrieving embeddings • Supabase Storage for handling uploaded files and generated content • Groq for LLM-powered generation and reasoning • Tavily for web-based research when web access is enabled • Memory system for retaining useful user-specific information • Tool-based agent execution with planning and execution stages 📚 Supported data & AI capabilities AIRW can work with multiple types of content, including: 📄 PDF documents 📊 CSV & Excel files 🎧 Audio files with transcription and timestamp-based retrieval 🖼️ Images using vision capabilities For documents, the system processes the content into smaller chunks, generates embeddings for those chunks, and stores them for semantic retrieval. When a user asks a question, AIRW retrieves the most relevant chunks. The system can also provide sources, allowing users to see where the information came from 🤖 It isn't limited to document Q&A either. AIRW supports: • Document summarization • Retrieval-based questions • General conversation • Web-assisted research • Memory-based responses • Python-powered analysis and visualization • Excel automation and modification • Multi-step tool execution through an agent workflow One of the things I particularly enjoyed was implementing many of these components myself instead of treating AI frameworks as black boxes. Understanding how chunking, embeddings, retrieval, planning, execution, memory, and tool orchestration work together has been one of the most valuable parts of this project. Github Repo: https://github.com/WaleedImran2007/AI-Research-Workspace Live Demo: https://airw.waleedimran.me Still learning. Still improving. 🚀 AI ArtificialIntelligence AIEngineering GenerativeAI RAG LangChain LangGraph FastAPI Python MongoDB AgenticAI MachineLearning BackendDevelopment FullStackDevelopment
I built Trust Check BD after losing 900 BDT to an online scam
In May 2026, I lost 900 BDT to an online scam. The money wasn't the biggest problem. The bigger problem was realizing that before sending money to a seller, there was no simple place where I could search and ask: “Has anyone reported this person or business before?” So I decided to build it. I started Trust Check BD — a free, community-powered platform where people can: → Search phone numbers, bKash/Nagad accounts, websites, Facebook pages, businesses, etc. → See reported scam cases → Submit reports with evidence → Help others avoid similar scams But building it wasn't just about creating a search box. Some of the engineering challenges were interesting: Normalizing Bangla + English search queries Matching different types of identifiers Connecting related reports without uncontrolled database crawling Evidence-based manual moderation Spam and abuse protection Redis caching for repeated searches Localized EN/BN SEO Automatically updating public report pages and sitemap data The most important lesson for me: When you're building a trust product, a false positive can be just as harmful as a missed scam. That's why a submitted report doesn't immediately become a public warning. It goes through validation, evidence collection, and human review first. I documented the full journey—from the original 900 BDT incident to the product architecture, challenges, technology choices, traction, and future roadmap. 👉 Full case study: monirhrabby.com/case-study/trust-check-bd I'm sharing this mainly to document what I learned while building a real product from a very personal problem. Search before you pay. Report when it happens. buildinpublic nextjs typescript mongodb prisma webdevelopment startup trustcheckbd
Shipped: TrackMyHunt – Job-Hunt Management Platform & Copilot
I built TrackMyHunt, a full-stack job-hunt management platform and companion Chrome extension built with React (Vite), Express.js, and MongoDB Atlas, deployed on Vercel and Render. The platform helps students and developers centralize their career search. Instead of managing fragmented spreadsheets, users can log applications, monitor pipeline stages (Applied, OA, Scheduled, Rejected), manage technical skill roadmaps, track resumes, and save study materials. The companion Chrome Extension (TrackMyHunt Copilot) allows users to extract and save job details (company, role, link, type) directly from job portals into the dashboard with a single click. The backend features stateless JWT authentication, password hashing with bcrypt, EmailJS-powered OTP verification, and Google OAuth. A reviewer can test and verify the acceptance criteria directly on the deployed application: - Registering for a new account with email OTP verification or signing in instantly via Google OAuth. - Navigating to the Application Tracker to log a new job application with specific hiring stages and job types. - Viewing the Dashboard to confirm that application metrics and stage counts update via MongoDB aggregation pipelines. - Editing an existing application's status (e.g., from Applied to OA) or deleting a record to verify full CRUD synchronization. - Testing the Skillboard, Resources Hub, Personal Notes, and Resume Manager to verify multi-section user data persistence. - Updating profile credentials (such as name or password) under account settings to confirm protected API route handling. - Installing the Chrome Extension in developer mode to test direct one-click job parsing and syncing from active tabs to the live backend. The complete source code and implementation are available in the project's GitHub repository, including the frontend UI, REST API micro-framework, database models, and Chrome extension scripts. Repository: https://github.com/ayushkhandelwal18/TrackMyHunt Live Web App: https://track-my-hunt.vercel.app/