// Software Engineer · AI agents & streaming systems

Software engineer for streaming and applied AI.

SDE II at Amazon Prime Video, working on streaming quality, test platforms and applied AI. Previously defense, national-lab and public-sector systems.

See the work GitHub Get in touch
5
years building production systems
270
device streaming lab
8
device families automated
63
tools in a production AI assistant

Selected work

Highlights from my current work.

Streaming deviceFire TV · Roku · Apple TVNetwork switchimpairment by policingClient telemetrywhat the app reportsHDMI capturelive video streamVision + timingframes → eventsMeasured QoETTFF · rebuffer · bitrateComparisoncause per disagreementreported
Streaming quality

Independent measurement of streaming quality

A system that measures what viewers actually experience (startup time, buffering, picture quality and errors) straight from each device's video output. It gives Prime Video an independent check on its own playback data.

  • Runs across every living-room, mobile and web device family
  • Findings led to changes in how playback quality is measured and reported
PythonComputer visionVideoAWS
Engineerasks a questionAgent loopLLM on BedrockRead-only toolsresults · logs · telemetryRun toolsrun test · watch live videoRemediationdiagnose → proposeTyped confirmation→ execute → audit → verifyhuman approves high-risk changes
Applied AI

AI assistant for test and lab operations

I designed the tool-calling architecture of my team's AI assistant on AWS Bedrock and built its testing, results and lab tools. Engineers run tests, inspect devices and diagnose failures in plain language, and nothing changes without their approval.

  • 63 tools across testing, results and lab operations
  • Approval-gated, fully audited actions
LLM agentsAWS BedrockTypeScriptPython
Web appReact · 25 routesSchedulercron jobs · device groupsAPI + LambdasPython · API GatewayLive videoin-browser device viewLab monitorpolls the lab every 10 minState storeDynamoDB · alarmsWarehouseRedshift stats · reports
Full-stack platform

Test operations platform

I'm the primary engineer on the platform my team uses every day to schedule tests, run a 270-device lab, review results and publish reports.

  • Wrote the majority of the codebase, front end to infrastructure
  • Serverless and container workloads on AWS
ReactTypeScriptPythonAWS CDKECS
Coordinator agentsees every threadAgent 1own workspaceAgent 2own workspaceAgent 3own workspaceAgent 4own workspaceShared knowledge basefacts · gotchas · decisionsDashboard + MCP toolslive state · overlap warningsharvest
AI-native engineering

Multi-agent engineering workflow

I run my day-to-day engineering through coordinated AI coding agents working in parallel, with a shared knowledge base so each new task starts from what earlier ones learned.

  • Parallel agents across model families
  • Live coordination dashboard
Multi-agentMCPCodexKiro
Test playbackads + live eventsPlayback datasessions · eventsAd + audience datalate-arriving rowsNightly validationfield-by-field checksNightly reportto data ownersFindingscause per mismatch
Ads & measurement

Ads and audience measurement quality

Automated nightly validation of audience-measurement and ad telemetry for Prime Video Advertising, so the data reported to partners can be trusted.

  • Nightly data-quality reporting to data owners
  • Every discrepancy traced to a cause
Ad techData qualitySQLJava

Before Amazon

Defense, national-lab and public-sector systems.

Inalab Consulting · 2024-2025

Lead engineer, classified DoD system

Led the upgrade of a Top Secret/SCI Department of Defense system as one of a two-person team, from design through delivery.

  • System-wide modernization in C++
  • Performance monitoring and tuning
C++Systems engineering

Application engineer, space data acquisition

Built satellite-data pipelines for a Department of Energy program, running on distributed compute the team built in-house.

  • High-volume data processing in Java
  • On-premises distributed computing
JavaDistributed systems
ICASA at New Mexico Tech · 2021-2023

Evidence system for a District Attorney

Built the evidence-submission system for the Second Judicial District Attorney's Office, adopted across Bernalillo County and drawing interest from other counties and states.

  • Worked directly with stakeholders up to the District Attorney
  • Promoted from intern to full-time
Java · Spring BootReact

Skills

What I reach for.

AI & LLM

LLM agents and tool use, multi-agent workflows, MCP, vision models.

Streaming & media

Playback quality, video pipelines, ad tech, audience measurement, TV and mobile devices.

Cloud & backend

AWS (serverless, containers, data), distributed systems, event-driven design.

Languages & frontend

TypeScript, Python, Java, Swift, Kotlin, C++, SQL; React.

Experience

5 years across streaming, defense, research and government systems.

2025 - now

Software Development Engineer II · Amazon Prime Video

Streaming quality, test platforms and applied AI. Promoted from SDE I; technical reference for partner teams; mentor.

2024 - 2025

Software Developer · Inalab Consulting

Lead engineer on a classified DoD system upgrade.

2023

Application Engineer · Sandia National Labs

Satellite-data pipelines on in-house distributed compute.

2021 - 2023

Software Developer · ICASA, New Mexico Tech

Evidence system for the Second Judicial District Attorney. Intern to full-time.

2022

B.S. Computer Science · New Mexico Tech