Senior Software Engineer @ SeekLab
September 2025 - Present
- ▶Architected production RAG system from ground up with embeddings pipeline, vector search (Pinecone), and retrieval evaluation—improved candidate match quality 35% and reduced time to hire 40% across enterprise accounts.
- ▶Drove end-to-end technical strategy for AI candidate matching engine; built consensus across Engineering, Product, and Data Science on semantic search vs. keyword-based approaches.
- ▶Engineered high-scale PostgreSQL infrastructure handling 10K+ concurrent lookups with 99.9% uptime; resolved N+1 patterns, connection pooling, read replicas—sub 100ms P95 latency under load.
- ▶Owned full-stack product development (NextJS and Node microservices) across 15+ features; established reusable component patterns, API contracts, and observability—accelerated throughput 30%, reduced production incidents 25%.
- ▶Established LLM productionization standards: prompt versioning, structured output schemas, function-calling; reduced hallucination rates 40% and enabled reliable A/B testing across recruiter workflows.
- ▶Designed fault-tolerant LLM workflows with retry logic, fallback strategies, and real-time monitoring; reduced AI feature failures 60% and established playbooks for debugging production LLM issues.
Technologies used:
NextJSNode.jsPostgreSQLPineconeOpenAILangChainRedisAWS