Backend / distributed systems engineer at Bank of New York Mellon building high-throughput event pipelines and reliable workflow-driven services (Kafka, gRPC, Redis, Temporal).

I solve performance + reliability problems — low-latency APIs, idempotency, retries, observability, concurrent systems — and build side-projects around distributed caches, task queues, developer tools, and real-time AI agents.

Currently exploring: storage systems, real-time architectures, MCP server development, and LLM integrations with strict structured outputs.

Static + dynamic analysis tool that finds goroutine leaks, deadlocks, and lock bugs in Go programs using execution traces — no instrumentation or code changes needed.

• 94% accuracy on GoBench GoKer — 64/68 real production bugs detected (Kubernetes, etcd, CockroachDB, gRPC, Docker)
• 0 false positives on net/http/httptest with 224 goroutines
• Goroutine provenance tree, Tarjan's SCC N-way deadlock detection, go/ssa static analysis, CI baseline flags

Voice-controlled AI agent that turns speech into system diagrams in real-time.

• Django + React + Groq LLaMA 3.3 + Deepgram
• Zero hallucination via Pydantic schema validation + Instructor
• ~4s end-to-end latency, 95%+ transcription accuracy

MCP server that gives AI agents (Claude Code, Cursor, Windsurf) a pre-build reality check — scans 6 sources with LLM-powered intent parsing before you write a line of code.

• Scans GitHub, HN, npm, PyPI, Reddit, Google/DDG in parallel
• Outputs competition_score + demand_score + gap analysis with dynamic insights
• Drop-in for any MCP-compatible client: claude mcp add prelaunch -- uvx prelaunch-mcp

High-performance distributed caching system with etcd service discovery, gRPC communication, consistent hashing, and Prometheus monitoring.

• 23,400 ops/sec with sub-1ms P50 latency (macOS M4)
• Thread-safe LRU cache with singleflight deduplication and concurrent access
• Hot data replication across nodes

Production-ready task queue with Redis, designed for reliability through at-least-once delivery.

• Concurrent worker pool with lease-based ownership and automatic recovery
• Priority queues, retry with exponential backoff, graceful shutdown
• Built-in Web UI dashboard for monitoring

Languages: Go · Java · Python · C++ · TypeScript

Systems & infra: Kafka · gRPC · Redis · etcd · Temporal · Postgres · Elasticsearch · Docker

AI & LLM: Claude API · Groq LLaMA · Deepgram · Instructor · Pydantic · MCP

Patterns: Consistent hashing · Singleflight · At-least-once delivery · Lease-based ownership · Exponential backoff · Goroutines · Worker pools · Thread-safe data structures

Building AI infra or real-time systems? Let's talk.