I lead engineering teams at Target while staying deeply hands-on — building production-grade AI agent systems, MCP integrations, and enterprise Java platforms. My work sits at the intersection of cutting-edge AI and shipping code that runs in production.

A multi-agent AI coding system: submit a spec, get production code with PRs opened against your repos. Spring Boot 3.3 + Spring AI 1.1 + React + PostgreSQL + Redis, deployed on GCP.

Pipeline: Smart Classifier → Outline/Skeleton Planner → parallel Task Expansion → Sequential Worker with native tool-calling → Reviewer (general + wiring passes) → auto-preview + GitHub PR.

What makes it interesting:

• Tri-provider native tool calling through one LlmRouter — Anthropic (Claude Sonnet/Haiku 4.x), OpenAI (GPT-5.4), Google Gemini (3 Pro/Flash). Per-provider circuit breakers, prompt-cache-aware cost tracking, BYOK + managed-key modes.
• Empirically validated — resolves SWE-bench Lite instances (medium 67%, hard 60% across 3 providers) and scores 98/100 on ProgramBench cmatrix.
• Multi-repo project workspaces with versioned architecture docs, RAG over project docs (pgvector), cross-repo contract registries, and a .executespec/ folder pattern that lives in customer git.
• Customer-facing surfaces — PR Review agent, Code Review scanner (semgrep + LLM triage), automated CI-failure patch mode, project chat with apply-suggestion, multi-tenant orgs + Stripe/Razorpay billing.
• Engineering discipline — ~180 Flyway migrations, 1,000+ test suite, ArchUnit guards, native tool-calling end-to-end (no XML parsing hacks), cleanroom Docker containers for safe code execution.

📬 Want a demo, architecture deep-dive, or to compare notes on agentic codegen systems? Email me or reach out on LinkedIn — happy to walk through the design.

Concurrent debates between OpenAI, Anthropic, and Gemini agents with a supervisor agent monitoring consensus in real-time. Spring Boot 3.4 + Spring AI + React + STOMP/WebSocket. Open source, runnable locally.

• project-chaos — Multi-agent debate system with OpenAI/Anthropic/Gemini and an automated supervisor for consensus detection
• trade-mcp-server — Stock trading MCP server with Spring AI
• mcp-client / mcp-server — Reference MCP client/server implementations
• zero-trade-app — Agentic AI trading on Zerodha Kite Connect with technical/news/research agents + risk manager
• fno-conservative-algo — Conservative F&O algorithmic trading with AI agents

• openai — OpenAI + Spring AI with RAG, pgvector, chat memory
• anthropic — Claude integration with Spring AI
• gemini — Gemini API with Spring AI
• javagpt — Lightweight local GPT in Java 21 + Deep Java Library (DJL)
• query-builder — Text-to-query generator

• telegram-bot-llm — Telegram bot with LLM code-gen
• telegram-bot — Telegram bot with LLM + MCP capabilities
• websocket — Simple chat over WebSockets

• drools ⭐ 5 — Drools Rule Engine + Spring Boot
• executor ⭐ 3 — Java Executor Framework patterns
• kafka — Kafka producer/consumer + Spring Boot
• elasticsearch — Elasticsearch + Spring Boot
• dms — Document Management System (MongoDB GridFS)
• bank-statement-analyser — Bank statement categorisation
• credit-risk-analyser — Credit risk from bank statement + bureau + GST/ITR
• jeasy — Jeasy Rule Engine + Spring Boot + Swagger
• stock-exchange-trade — Groww API wrapper for stock & F&O
• fno-dashboard — Investment & strategy tracking UI
• cowin — CoWin Public API integration

• montecarlo ⭐ 1 — Monte Carlo simulations
• geneticAlgo — Genetic algorithm implementation
• tictactoe — Tic-tac-toe with minimax AI player
• network — Mail-train control system (Java)

Specialties: AI agent orchestration · multi-LLM routing & failover · Model Context Protocol (MCP) · RAG + pgvector · prompt caching · rule engines (Drools, Jeasy) · distributed systems · real-time messaging (Kafka, STOMP, SSE) · observability (Micrometer, OTel, Prometheus, Tempo, Grafana) · algorithms (Monte Carlo, Genetic, Minimax, Dijkstra)

• Production patterns for multi-agent codegen pipelines (planner → worker → reviewer with native tool calling)
• Cost-aware LLM routing with prompt-cache hit-rate optimisation across Anthropic / OpenAI / Gemini
• SWE-bench-style empirical evaluation of agentic systems on real-world bug-fix benchmarks
• High-performance LLM inference with Java + DJL

Open to chats on AI/LLM in enterprise Java, Spring AI + agent orchestration, MCP servers/clients, rule engines, distributed systems, and algorithmic problem-solving. If you're building something agentic and want to compare notes — or want to learn more about ExecuteSpec — drop me a line.

• 💼 LinkedIn
• 📧 navneet.prabhakar007@gmail.com
• 🐙 You're already here — explore the repos above

💡 Building AI agent systems that actually ship to production.