I’m a senior full-stack software engineer who enjoys building scalable, reliable systems that hold up in real production environments.

Elixir · Rust · Python · Ruby · Distributed Systems · React · TypeScript · AWS · AI/ML · LLM · RAG · Vector Search

I design and build backend systems that handle high concurrency, unpredictable traffic, and real-world failure modes. That includes event-driven architectures, background processing, and APIs that stay fast and stable as usage grows.

Much of my recent work involves AI-powered product systems, where backend services integrate LLMs, RAG pipelines, and vector search to deliver reliable AI features in production. These systems often require careful design around latency, evaluation, and guardrails so model behavior stays predictable as usage scales.

On the frontend, I focus on clean, maintainable interfaces—often real-time or data-heavy—where performance and user experience matter just as much as correctness. I enjoy bridging the gap between backend systems and frontend behavior so the whole stack works coherently, especially when building AI-assisted workflows and interactive AI features.

• Prefer simple, explicit designs over clever abstractions
• Optimize for observability and debuggability, not just raw speed
• Treat performance, reliability, and developer experience as first-class concerns
• Expect systems to fail, and design them to fail predictably
• Apply the same discipline to AI systems, where evaluation, guardrails, and monitoring matter as much as model capability

I’m most interested in problems that only show up at scale: latency spikes, backpressure, race conditions, and systems that work perfectly—until they don’t.

That includes AI systems in production, where model behavior, prompt drift, and data quality can introduce entirely new failure modes.

In recent years I’ve focused heavily on building AI-enabled product systems and LLM-powered features in production environments.

Experience includes:

• Building LLM-powered product features using OpenAI APIs and Python services
• Designing Retrieval-Augmented Generation (RAG) pipelines grounded in internal knowledge bases
• Implementing vector search and semantic retrieval systems
• Developing AI experimentation frameworks for evaluating model quality and product impact
• Designing AI guardrails, rate limiting, and observability for LLM APIs
• Building analytics pipelines to collect user feedback and prompt evaluation data
• Running A/B testing and offline evaluation pipelines for AI features
• Scaling AI services supporting millions of AI-powered requests per month

• High-concurrency backends and distributed workflows
• Data pipelines and background job systems
• Real-time applications and event streams
• Cloud infrastructure and production operations
• Frontend performance and long-lived codebases
• Production AI systems, RAG architectures, and LLM integrations

I enjoy leaving systems clearer, calmer, and easier to operate than when I found them.

Quietly shipping production software.