Matthew Powers
I’m Matthew Powers, a senior AI software engineer with 11+ years of experience building production-grade distributed systems, ML infrastructure, and generative AI applications.
About
I’m Matthew Powers, a senior AI software engineer with 11+ years of experience building production-grade distributed systems, ML infrastructure, and generative AI applications. I focus on taking models and AI agents from experimentation to reliable deployments—especially through LLM orchestration, retrieval-augmented generation (RAG), model evaluation, scalable inference, and observability.
In my recent work, I’ve led the development of AI-powered search and agent systems that combine large language models with real-time retrieval and tool execution. Earlier, I built large-scale machine learning infrastructure for LLM experimentation, post-training, and inference, and I started my career strengthening backend foundations through distributed observability and full-stack engineering. I enjoy architecting robust APIs and pipelines, improving throughput and latency, and ensuring quality and regression safety as AI systems evolve.
- 11 years' experience
- Joined November 2026
Industry experience
- Software & Internet
- Computers & Electronics
- Professional Services
More experts in Cheyenne, United States
Work experience· 4 roles
Led development of AI-powered search and agent systems that combine large language models with real-time information retrieval, semantic search, retrieval-augmented generation, ranking, and external tool execution. Architected Python services for LLM orchestration, retrieval, inference, and automated evaluation to support scalable, low-latency production workloads. Developed agentic AI infrastructure for multi-step research, tool interaction, information synthesis, and grounded response generation by connecting model capabilities with retrieval and application services. Provided technical leadership across AI architecture, distributed systems, performance optimization, system reliability, and engineering implementation.
Built large-scale machine learning infrastructure supporting LLM experimentation, model evaluation, post-training, and inference workflows across distributed compute. Developed Python and PyTorch pipelines for data processing, experimentation, and automated evaluation while connecting research workflows to reproducible production infrastructure. Created evaluation systems to measure model quality, regressions, reliability, and behavioral performance across model versions and experimental configurations. Collaborated with research and infrastructure engineers to productionize experimental ML techniques while improving throughput, observability, reproducibility, and infrastructure efficiency.
Developed backend and application components for distributed observability systems processing high-volume metrics, logs, traces, and telemetry. Built Go and Python services plus REST and gRPC APIs, along with Kafka-based processing pipelines and React/TypeScript interfaces for monitoring and analytics workflows. Designed distributed data-processing components and optimized databases and caching layers to improve throughput, latency, and production reliability. Contributed to Kubernetes deployments, CI/CD, testing, production observability, and incident response for increasingly complex distributed workloads.
Education
Hire Matthew Powers today
Post a job and invite Matthew, or contact them directly. No hidden commission.