Senior AI Platform Engineer
CodeGPT
Jul 2023 - Current (3 years 4 months)
Led backend architecture for an AI platform delivering real-time code analysis and generation tools to engineering teams. Designed and implemented multi-tenant FastAPI services supporting high-volume code ingestion, contextual reasoning, and LLM-based inference. Built production RAG pipelines combining vector search and LLM reasoning, improving suggestion accuracy by ~28% and reducing irrelevant outputs. Developed orchestration workflows for multi-step reasoning tasks, ensuring deterministic validation and improving output reliability. Reduced average inference latency by ~40% through asynchronous processing, task queues, and autoscaling optimizations. Implemented monitoring, tracing, and alerting systems that reduced incident resolution ti