AI Engineer focused on agentic AI systems, multi-agent orchestration, and production-ready backend architecture.
- Building reliable LLM-powered workflows for real business use cases
- Interested in scalable AI systems, RAG pipelines, and real-time applications
- Based in Dhaka, Bangladesh
- Email: zaidrehman3103@gmail.com
- LinkedIn: zaidrehmann
- Google Scholar: Profile
- GitHub: zaidrahmann
- AI Solutions Engineer Intern at SJ Innovation
- Designing cost-efficient LLM systems (token efficiency, routing, memory distillation, prompt caching)
- Building agentic automation pipelines with n8n, webhooks, and REST APIs
- Languages: Python, JavaScript, Java, C++, SQL
- AI/ML: PyTorch, deep learning, CNNs, embeddings, retrieval and reranking
- LLMs & Agents: RAG, LangChain, LangGraph, LangSmith, Google ADK, tool-calling agents
- Backend & Web: FastAPI, Node.js (Express), REST APIs, WebSockets, React, Zod
- Databases/Search: PostgreSQL, MongoDB, Drizzle ORM, FAISS, vector search
- Tools: Git, Docker, Postman, n8n, Zapier
Multilingual RAG system built from scratch with Python, FAISS, and sentence-transformers, with reranking for higher answer precision and FastAPI + Gradio delivery.
Full-stack real-time match tracking platform using Node.js, PostgreSQL, Drizzle ORM, React, and resilient WebSocket architecture.
MERN-based AI diet planner with personalized recommendations and LLM API integration.
- Privacy-Aware Feature Envy Detection through Survey-Driven Federated Learning
- Modern Imaging and Learning Methods for Burn-Depth Assessment: A Structured Review
If you want to collaborate on AI systems, backend engineering, or research-driven projects, feel free to reach out.