Staff Engineer at Netflix
I build production infrastructure for autonomous AI agents.
I design the control-plane layer that lets large engineering organizations safely give agents more authority: secure runtimes, model and tool observability, cost attribution, policy enforcement, eval-backed reliability, and targeted kill switches.
Founding member of Netflix's platform modernization team. I design and operate production agent infrastructure across coding, Slack, research, and support workflows — including systems that have shipped thousands of validated agent-driven pull requests.
Founding Member
Netflix Platform Modernization
Thousands of PRs
Validated agent-driven changes shipped
Agents in Production
Coding, Slack, research, and support workflows
Control Plane Focus
Runtime safety, observability, cost, policy, and evals
Production Agent Infrastructure
Designing the control-plane layer that lets large engineering organizations safely operate autonomous AI agents — runtime, observability, cost, policy, and evals.
Agent Control Plane
The infrastructure layer for operating autonomous AI agents at enterprise scale.
Secure Agent Runtime
Sandboxed execution, least privilege, and bounded authority for production agents.
Autonomous Software Evolution
Production agents shipping thousands of validated pull requests at Netflix scale.
Side projects and long-running interests
Work outside my main focus on production AI agent infrastructure — kept here for range, not for emphasis.
Capital & Real Estate
Disciplined capital allocation and real-world operations.
Brain-Inspired AGI
Exploring intelligence through temporal memory, cortical columns, and local learning.
Fiction
Sci-fi writing and speculative world-building.
Busy Beaver
Interactive explorations of foundational computer science.
LazyFish
Privacy-first browser utilities — no uploads, no tracking.
Operating Autonomous Agents
A series on the infrastructure required to run AI agents as production systems: control planes, gateway observability, runtime safety, cost governance, evals, memory, and kill switches.
The Agent Control Plane
Autonomous AI agents are becoming a new production substrate. The hard enterprise problem is not just building smarter agents; it is operating them with identity, observability, policy enforcement, cost attribution, runtime isolation, evals, and kill switches.
Agent Observability Belongs at the Gateway
Client-side logs are not enough for enterprise AI agents. The model gateway is the best choke point for observing model calls, tool intent, cost, risk, and agent behavior.
AI Cost Control Is an Engineering Discipline
AI spend will not be controlled by asking engineers to use models less. It needs engineering systems: attribution, model routing, prompt caching, context management, budgets, evals, and outcome-based measurement.
Get in touch
Open to conversations about agent infrastructure, AI control planes, secure runtimes, model gateways, AI cost governance, technical architecture, speaking, and selective collaborations.