Hire Sanjay J. - Forward Deployed Engineer

Partnered directly with customers to translate business workflows into production-grade agentic AI systems and iterating rapidly from prototype to launch, acting as the primary technical communicator for both technical and non-technical stakeholders - Built end-to-end conversational agents that collected user inputs across text + images, validated/structured data, and automated downstream decisioning and handoffs to external stakeholders - Implemented autonomous web-based automation using Playwright to complete critical workflows in third-party partner portals when APIs were unavailable (form filling, configuration, verification) - Designed lead ingestion and enrichment pipelines that fetched data from external sources, normalized it into structured fields, computed propensity-to-buy scoring, and routed prioritized leads to human sales teams - Created internal tooling to coordinate multi-agent workflows, manage state, and enforce guardrails/retries across long-running, multi-step tasks - Built observability and debugging capabilities for agent execution (step-level traces, decision explanations, failure surfacing), improving reliability and enabling customer-friendly transparency - Deployed and operated systems on Google Cloud (compute/storage), selecting and integrating best-fit foundation models per use case (OpenAI, Claude, Gemini) with cost/latency/quality tradeoff analysis

* Leveraged Generative AI to build applications accelerating clinical research: complex medical writing copilots, medical text chatbots, and dynamic NER pipelines * Acted as Lead on Customer Engineering team, liaising directly with biotech/pharma customers to gather requirements and build products fit for their use-case * Implemented RAG applications using LangChain, LlamaIndex and vector stores like ChromaDb and PostgreSQL Pgvector to construct accurate and reliable clinical research copilots * Implemented cutting edge techniques and datastores to chunk, embed, store and retrieve data with high accuracy and flexibility on model selection; Knowledge Graphs, RAPTOR, StructRAG, Contextual Retrieval * Developed highly resilient data pipelines capable of large throughput using Dagster as an orchestration tool in concert with Kubernetes to manage scale * Deployed applications using Helm charts with Kubernetes on Microsoft Azure cloud infrastructure