Technical Lead | AI Agent Platforms and Distributed Systems | Python, Java
Pleasanton, CA | dipankar.banik@gmail.com | (201) 736-5586 | linkedin.com/in/dipankarbanik | github.com/dipankarbanik
dipankarbanik.com | US permanent resident, no sponsorship needed
Technical lead with 20+ years in production distributed systems, now applied to LLM-powered agents: agent harnesses, context engineering, evaluation, and governance. Ships agent workflows to production with MCP tools, RAG, and guardrails around the model and a policy engine making the final, auditable decision. Builds platform capabilities other teams depend on, sets engineering standards, and leads teams of six to ten across the US and India.
In production at Narvar
Independent projects, 2025 - 2026 (Python, end to end)
Senior Backend and Platform Engineer | Narvar Inc., San Francisco, CAMar 2016 - Present
Applications Engineer | Gilead Sciences, Foster City, CASep 2014 - Mar 2016
Portal Lead | Cisco, San Jose, CAAug 2010 - Aug 2014
Earlier Experience | Lead Portal Architect and Java EE Developer, NY, NJ, and CA2001 - 2010
Languages: Python (FastAPI, asyncio, Pydantic), Java 21 (Spring Boot 3), Go, SQL, TypeScript
Agents and LLMs: LangGraph, LangChain, MCP, multi-agent orchestration, human-in-the-loop, OpenAI and Anthropic APIs, model routing, caching, cost control
Context, evaluation, safety: RAG, vector databases, embeddings, LLM-as-judge, regression tests in CI, guardrails, prompt-injection defense, PII redaction, OpenTelemetry
Platform: event-driven architecture, Apache Pulsar, Kafka, REST, gRPC, SSE, PostgreSQL, Redis, BigQuery, Docker, Kubernetes, Helm, Terraform, GCP, AWS
Security: OAuth 2.0, JWT, Okta, mTLS (Linkerd), HashiCorp Vault, Google Tink, ModSecurity WAF, policy engines (Drools)
Leadership: teams of six to ten across the US and India, mentoring, engineering standards, design and incident reviews
B.S. in Physics, University of Pune, India