Jawad Khan

I’m a backend & AI engineer with 5.5+ years building production-grade distributed systems and LLM-powered applications at a technical-lead level. I set architecture standards, raise the technical bar through code reviews, and guide delivery across concurrent systems.

I architected NIAMR Pakistan’s 24-service, event-driven microservices platform—building 10 AI model services that process live TV streams for real-time monitoring. At BeatCodes, I led the design of a scalable Django + Kafka architecture for real-time media ingestion, orchestrating 120+ channels for guaranteed delivery.

On the agentic AI side, I’ve shipped production multi-agent systems with LangChain + LangGraph—using tool routing, stateful graph execution, and prompt-injection guardrails. I integrate HITL checkpointing so database writes require explicit approval, and I use LangFuse for end-to-end observability (tracing, token usage, latency, and prompt versioning).

I also build for reliability and scale: Kafka media streaming with 8 GB+/minute throughput, Elasticsearch for sub-second search across 10M+ assets, and CI/CD quality gates with migration-safety checks. I’ve published NLP research in 2023 and I’m motivated by systems work that turns real-time data and robust AI tooling into dependable products.