Open to opportunities

Michael Chen

@michaelchen2

Message

Senior AI engineer building reliable, production-grade LLM and RAG systems.

United States

Message

What I'm looking for

I seek roles building reliable, scalable LLM/RAG systems with strong MLOps practices, safety-first releases, and cross-functional impact in regulated or high-traffic environments.

Senior AI Engineer with 7+ years of experience building production-grade ML and LLM systems across healthcare, enterprise, and high-growth startups. Deep expertise in Agentic AI, RAG architecture, and LLM reliability — including hybrid retrieval, multi-agent orchestration, provider fallbacks, evaluation-first releases, and hallucination mitigation. Proven track record of delivering high-traffic, regulated systems on AWS and Kubernetes, with strong discipline in MLOps, CI/CD, observability, and safe rollout. Known for designing AI systems that hold up under real-world conditions: latency spikes, partial failures, audits, and continuous iteration.

Experience

Work history, roles, and key accomplishments

Current

Senior AI Engineer

Current

Abridge

Dec 2024 - Present (1 year 2 months)

Architected and deployed agentic clinical automation and Agentic RAG pipelines to reduce hallucinations and improve grounding, while delivering high-traffic LLM services on AWS EKS with autoscaling and graceful degradation for latency spikes.

Python LangGraph Agentic RAG Terraform GitHub Actions Prometheus Grafana

Senior Machine Learning Engineer

Deloitte

Nov 2022 - Nov 2024 (2 years)

Built enterprise RAG and stateful agent services, operationalized evaluation-first release gates and progressive GitOps rollouts, and ensured reliability via caching, provider fallbacks, and observability across AWS EKS deployments.

Python fastAPI LangGraph RAG Terraform ArgoCD Redis GitHub Actions

Machine Learning Engineer

Cohere

Sep 2020 - Oct 2022 (2 years 1 month)

Developed transformer-based models, distilled and quantized models for improved serving throughput, and owned high-traffic inference and semantic search pipelines combining hybrid retrieval for improved relevance and robustness.

Python PyTorch Transformers Model Distillation Model Quantization Semantic Search AWS

AI Engineer

Palantir Technologies

Jul 2018 - Aug 2020 (2 years 1 month)

Built PySpark forecasting and anomaly-detection pipelines and productionized ML services on Kubernetes with autoscaling, drift detection, and CI/CD to maintain reliability under operational constraints.