Open to opportunities

Evan Thomas

@evanthomas1

Message

Senior AI/ML engineer specializing in production LLM systems, observability, and reliability.

United States

Message

What I'm looking for

I seek roles where I can own production LLM systems—observability, evaluation, and inference—within collaborative teams focused on reliability, measurable product impact, and scaling real-world AI deployments.

I am a Senior AI/ML Engineer with 11+ years of experience and 6+ years building and operating production ML and LLM systems, owning end-to-end architecture, launch, scaling, on-call operations, and multi-year iteration informed by real-world usage.

I specialize in LLM observability, evaluation, agent reliability, and inference infrastructure, with hands-on experience designing tracing and evaluation pipelines, human-in-the-loop labeling workflows, drift detection and alerting, autoscaling GPU-backed inference, incident response, and partnering with product and enterprise customers to translate quality issues into measurable system improvements.

Experience

Work history, roles, and key accomplishments

Senior AI Engineer

Arize AI

Jan 2024 - Jan 2026 (2 years)

Owned core components of a production LLM observability platform ingesting millions of traces/day; designed tracing, offline and online evaluation pipelines, and human-in-the-loop workflows that reduced undetected regressions and improved issue resolution. Served on-call for ingestion, cost, and evaluation incidents and partnered with customers to translate quality complaints into measurable syste

LLM Observability Évaluation Human In The Loop Python Distributed Systems Incident Response

Senior AI Engineer

Galileo AI

May 2021 - Jan 2024 (2 years 8 months)

Built observability tooling for LLM-powered agents capturing multi-step traces and implemented failure taxonomies, adaptive sampling, and guardrails that produced double-digit improvements in task completion and reduced high-severity failures. Created dashboards and analysis workflows to surface recurring production failure patterns.

Failure Taxonomy Adaptive Sampling GuardRails Dashboards Reliability Engineering Python

Senior Machine Learning Engineer

Baseten

Mar 2018 - May 2021 (3 years 2 months)

Designed and optimized GPU-backed real-time and batch inference pipelines to meet strict p95 latency targets, implemented autoscaling strategies and observability for latency and resource utilization, and participated in incident response for inference outages and performance regressions.

AutoScaling Observability Kubernetes Python Incident Response

Software Engineer

Google

Oct 2014 - Mar 2018 (3 years 5 months)

Built and maintained large-scale backend services and user-facing features in high-traffic production systems, participated in on-call rotations and postmortems, and contributed to long-term reliability and distributed systems practices used by millions of users.