AI & full-stack engineering

The work. Open to inspection.

Explore the complete portfolio: production systems, open-source engineering and commerce experiences. Read a detailed case study or inspect the original project and contribution sources.

Official OpenAI Agents SDK launch artwork showing agent handoffs and guardrails on a blue abstract background

OpenAI Agents SDK / Open-source contribution

OpenAI — engineering the official Agents SDKs

Merged upstream · Python & JavaScript

Strengthening the infrastructure behind production AI agents: fail-fast validation against ambiguous tool dispatch, collision-resistant MCP tool names and application-controlled UI streaming. Implemented, regression-tested and merged into OpenAI’s official Python and JavaScript SDKs.

Read the case study
Stanford University production software case study on structured sentiment analysis

Stanford University / Selected production software delivery

Stanford University — AI for a global research leader

$2.3B in FY2025 sponsored research support, including SLAC

Selected production software delivery for Stanford University. Structured sentiment analysis, target extraction and evidence-grounded outputs, with abstention rules that keep unsupported inferences out of the decision path. The public technical note explains the architecture.

Kairaxis marketplace homepage showing intent-led discovery across AI skills, prompts and workflows

Kairaxis / Client production system

Kairaxis — a marketplace for the agent economy

MCP-ready · Public APIs · Client production platform

Architected and delivered a live AI marketplace for a client: searchable skills, prompts and workflows, inspectable provenance, permissions and compatibility, plus public APIs and authentication. A full-stack product built for people and agent clients.

AgentrixOS homepage showing an all-systems-online interface for autonomous business agents

AgentrixOS / Client production system

AgentrixOS — OpenAI, Anthropic & Google in one platform

Model integrations · Client production platform

Architected and delivered a client platform that brings multi-model AI agents into business operations. Orchestration, RAG, knowledge connectors, BYOK and guardrails form one system for deploying, supervising and integrating specialized agents.

An international engineering practice

Bring the hard problem. Let’s build the way through.

Tell me what needs to work, what is blocking it and what success looks like. I’ll review the fit and help define the next move.

Discuss your project