Meet Merchant

About

At Esri, the candidate is building a production LLM-powered WebGIS AI framework that embeds an AI assistant into web maps and routes natural-language requests into ArcGIS JavaScript SDK functions via a LangGraph-orchestrated, multi-agent system. They emphasize production reliability and scale (strict tool calling/JSON, live schema validation, query guardrails) and rigorous evaluation/observability using LangSmith, offline prompt datasets, and latency/tool-call accuracy tracking.

Key Strengths

• Built and deployed an LLM-powered WebGIS assistant framework integrated with ArcGIS JS SDK
• Designed modular multi-agent architecture with extensibility for third-party custom agents
• Strong reliability engineering for LLM systems (strict tool calling, live-state validation, guardrails)
• Production-grade observability and evaluation (LangSmith tracing, offline evals, accuracy/latency tracking)
• Cost/performance optimization via model comparisons and selective use of cheaper/faster models
• Token and context optimization using embeddings + cosine similarity to shortlist relevant fields
• Cross-functional collaboration with GIS product/domain expert and incorporation of user feedback

Experience

Education

Awards

• Principal’s Award - University of Mumbai
• Above and Beyond - JPMorgan Chase & Co.
• Out of the Box Thinker - JPMorgan Chase & Co.

Skills

Languages

Certifications