.png&w=256&q=75)
syed mehrab@syedmehrab
1
Events attended
1
Submissions made
Pakistan
6+ years of experience
About me
Hi, my name is Syed Mehrab. I’m a Python backend and AI engineer with around four years of experience building backend systems and AI-powered applications. In my recent work, I’ve been focusing a lot on Large Language Models and building production AI systems like chatbots and conversational platforms. I’ve worked on systems using RAG pipelines where we retrieve knowledge from documents and combine it with LLMs to generate accurate responses. On the backend side, I usually work with Python frameworks like Django and FastAPI. I’ve also built microservices using gRPC with streaming and deployed AI models using tools like BentoML and vLLM on GPU servers. Recently I’ve also been working on AI automation workflows and early-stage agentic systems where AI can perform tasks and interact with external tools. Overall, I really enjoy working on real-world AI products, especially where backend engineering and AI systems come together.
🤓 Latest Submissions
LearnArc – Pay-Per-Use AI Learning Platform
LearnArc is a pay-per-use AI education platform designed to make learning more flexible and cost-efficient. Instead of paying upfront for full courses, users are charged micro-amounts based on actual consumption — per minute of video watched, per page of PDF read, and per AI tutor interaction. The platform uses a modern full-stack architecture with a Next.js frontend deployed on Vercel and a FastAPI backend hosted on Render. MongoDB is used for storing users, courses, lessons, transactions, and chat history. The backend follows a clean layered architecture (Controller → Service → Repository) to ensure scalability and maintainability. An integrated AI tutor enhances the learning experience by providing real-time assistance using LLM APIs. Every interaction is tracked and priced dynamically, enabling a true usage-based billing model. The system also includes authentication with JWT, transaction tracking, and a wallet system that aggregates user spending. Although blockchain payments are currently mocked, the architecture is designed to easily integrate real micropayments such as USDC in the future. This project demonstrates how AI, microtransactions, and modern cloud deployment can be combined to build the next generation of accessible and fair digital education platforms.
26 Apr 2026
