Christopher Liu
@christopherliu1
Senior software engineer focused on secure AI infrastructure, distributed systems, and reliable production backends.
United StatesWhat I'm looking for
I’m a Senior Software Engineer with experience spanning AI infrastructure, distributed systems, and product security across Anthropic, Airbnb, Twitch, and Facebook. I’ve led platform workstreams for MCP-backed agent tooling, AI customer support automation, high-scale notification fanout, and runtime privacy enforcement—building systems designed to be reliable, secure, and scalable.
I specialize in backend and systems engineering with TypeScript, Python, Go, React, and AWS, alongside deep operational expertise in observability and policy-aware execution. From policy-aware MCP gateways and LLM workflow tracing to evaluation/replay tooling and runtime invariant enforcement, I bring a production-first mindset: measure everything, debug fast, and ship safe AI experiences that connect advanced product capabilities to robust infrastructure.
Experience
Work history, roles, and key accomplishments
Led a defined workstream for MCP-backed tool execution across Claude Code and Claude API agent workflows, improving how coding agents connect to external systems like GitHub, Jira, Slack, and Postgres-backed tools. Built a policy-aware MCP gateway with per-tool allowlists, credential-scoped execution, and audit logging, and delivered TypeScript/Python backend services to trace and debug multi-step
Led the backend/platform workstream for a reservation-aware AI customer support assistant for guests and hosts across cancellation, refunds, booking changes, and payments. Built bounded-context orchestration and a RAG grounding layer, implemented model-agnostic LLM orchestration with escalation routing for high-risk cases, and shipped evaluation/replay tooling in Python/FastAPI plus internal React
Led the backend notification workstream for Twitch Stories, designing event contracts, fanout flow, preference evaluation, and delivery tracking to notify followers when creators publish new stories. Built Go-based fanout services on Amazon ECS and DynamoDB with eligibility logic, and implemented idempotency, retry, TTL cleanup, and dead-letter handling while adding end-to-end observability for in
Built and operated real-time distributed notification systems delivering hundreds of millions of messages per day across push, web, in-app, and email-style channels. Developed Go-based services on Amazon ECS and DynamoDB to ingest events, expand audiences, enforce preferences and suppression logic, and ensured at-least-once delivery reliability with idempotency keys, retry-safe state transitions,
Software Engineer
Jul 2017 - Jan 2019 (1 year 6 months)
Worked on Facebook Product Security’s Invariant Detector, a runtime defense system that learned authorization invariants from production traffic and detected anomalous graph mutations before privacy bugs could be exploited. Contributed to live request-path invariant checking, improved enforcement safety with staged rollouts and violation classification, reduced false positives with better filterin
Education
Degrees, certifications, and relevant coursework
Carnegie Mellon University
Bachelor of Science, Mathematics and Computer Science
2013 - 2017
Earned a B.S. in Mathematics and Computer Science from Carnegie Mellon University (2013–2017).
Tech stack
Software and tools used professionally
Availability
Location
United StatesAuthorized to work in
Job categories
Skills
Interested in hiring Christopher?
You can contact Christopher and 90k+ other talented remote workers on Himalayas.
Message ChristopherFind your dream job
Sign up now and join over 100,000 remote workers who receive personalized job alerts, curated job matches, and more for free!
