utpal@portfolio:~$

$ whoami

Utpal Das

$ role --current

GenAI Engineering Manager, Deloitte India

Architecting RAG, knowledge-graph, and multi-agent systems that go to production at Fortune 500 scale.

$ facts --headline

✓ 10+ years shipping software, from Aetna's policy UI to Allianz's agent fleets

✓ 4 GenAI platforms in production at Pfizer & Allianz

✓ Still writes code every night — the projects below are the proof

View the work → Download résumé Say hello

$

$ experience --timeline --reverse-chronological

  1. Mar 2026 — Present Client: Allianz

    Deloitte India · Engineering Manager, Technical Program & Architecture Lead

    Agentic Data Migration Platform

    • Architected and led delivery of a 6-agent Claude Code system — clone, repo analysis, dataset lookup, code rewrite, output verification, feature-branch push — automating Foundry-to-Databricks migration end to end.
    • Cut delivery time up to 90% and migration cost up to 95% versus manual migration, while raising output per engineer.
    • Reduced per-run token cost with a terse-prompting ("Caveman") strategy and shipped an end-of-run report tracking assets migrated and tokens spent.
    • Integrated the pipeline across five external systems via MCP: Foundry, Databricks, Atlassian, GitMCP, Synapse.
  2. Jul 2021 — Mar 2026 Client: Pfizer

    Deloitte India · Engineering Manager, GenAI Platform Lead

    Oncoscout · Mapii · Constellation

    • Oncoscout — knowledge-graph chat over Neo4j/Cypher with single- and multi-hop retrieval, synonym lookup, and multi-modal SharePoint ingestion. Owned engineering execution and resolved a Cypher performance / hybrid-RAG stability issue that protected the launch timeline.
    • Mapii — shipped configurable chat and temperature settings for Pfizer Oncology, then led cross-team evaluation of clusterRAG/graphRAG as the next retrieval architecture.
    • Constellation — an enterprise RAG platform for Q&A, search, and summarization. A RAGAs-based evaluation gate cut hallucination-flagged responses by an estimated 30%. Negotiated production scope (SSO, audit logging) directly with client stakeholders. Ran on AWS EC2/Docker at an estimated 99.5%+ uptime, ~2,000+ queries/day.
  3. Jun 2019 — Jun 2021 Client: BrightEdge

    Mindfire Solutions · Senior Python Developer

    Spark (SEO platform)

    • Built crawler/parser infrastructure processing an estimated millions of pages a month across 50+ client domains to feed SEO ranking recommendations.
    • Shipped "Save for Later" bookmarking, adopted by an estimated 10,000+ users.
    • Architected a Google Maps → Leaflet.js migration saving an estimated $50K+ a year.
  4. Jul 2018 — May 2019 Client: Walmart

    Cognizant · Python Developer

    Partnerbot

    • Built an NLP/NLU seller-support bot (RASA) and a Flask template-generation service that standardized onboarding data (XML/Excel/JSON) for Walmart Marketplace sellers.
  5. Oct 2015 — May 2018 Client: Aetna

    Infosys · Systems Engineer

    PAR UI

    • Designed a MERN-stack policy/agent lookup application and the regression test framework that became the team's automated QA baseline.

$ ls ./projects

./enterprise client work — architecture & impact, code under NDA

Agentic Data Migration Platform

Allianz

Six orchestrated Claude Code sub-agents automate a full Foundry-to-Databricks migration. Up to 90% faster, up to 95% cheaper than doing it by hand.

Claude Code · MCP · Databricks · Foundry

Oncoscout

Pfizer Oncology

Knowledge-graph chat with single- and multi-hop retrieval, synonym lookup, and multi-modal document ingestion.

Neo4j · Cypher · Hybrid RAG

Constellation

Pfizer

Enterprise RAG platform for Q&A, search, and summarization. RAGAs eval gate, ~2,000+ queries/day at ~99.5%+ uptime.

FastAPI · LangChain · AWS · Docker

Mapii

Pfizer Oncology

Configurable chat and temperature controls, plus the evaluation that got clusterRAG/graphRAG adopted as the next retrieval architecture.

LangChain · GraphRAG · ClusterRAG

./personal nights & weekends — code is public

Agent Ops — Mission Control

Jira for AI agent fleets: natural-language briefs become tasks, a routing matrix assigns the model by criticality × size, and a kanban board tracks them live.

Node.js · Claude Code · LLM routing

Live app ↗ Code ↗

Rock · Paper · Scissors — 3D Multiplayer

Real-time 3D multiplayer, built on Three.js with zero build step. Procedural geometry, synced countdown, a small WebSocket relay for cross-device play.

Three.js · WebSockets · Web Audio

Live app ↗ Code ↗

ascii_image

Drop an image in, get ASCII art out — entirely in the browser, no server round-trip.

JavaScript · Image processing

Live app ↗ Code ↗

pyStocks+

A BUY/SELL/HOLD signal derived from live NSE promoter-shareholding trends, with manual fallback when NSE rate-limits a lookup.

Python · Flask · NSE data

Code ↗

$ cat skills.json

{
  "ai_llm": ["RAG", "GraphRAG", "Hybrid RAG", "Knowledge Graphs", "Neo4j", "Cypher", "LangChain", "LangGraph", "Multi-Agent Orchestration", "Claude Code", "MCP", "LLM Evaluation (RAGAs)"],
  "platform": ["Python", "FastAPI", "React", "TypeScript", "AWS", "Databricks", "Postgres", "Docker"],
  "leadership": ["Technical Program Management", "Solution Architecture", "Cross-Functional Delivery", "Stakeholder Negotiation", "Risk & Launch Readiness"]
}

$ whoami --credentials

certifications

  • Google Certified GenAI Developer
  • Neo4j Certified Graph Data Science Developer
  • HackerRank Certified Python, Java, SQL Developer
  • Deloitte Certified Project Manager

education

  • B.Tech, Electrical & Electronics Engineering
    Siksha 'O' Anusandhan University · 2011 – 2015

$ contact --me

Hiring for GenAI, platform, TPM, or architect roles — or just want to talk knowledge graphs and chess? My terminal's always open.

utpaldas.6osm@gmail.com --linkedin --github --twitter --youtube --resume