Peter Wang

Schweitzer Engineering Laboratories

Pullman, WA

• Architected three C#/.NET services over RabbitMQ, TCP, and REST automating selective solder program selection; cut cycle time 18% (+60 boards/day), stable in 24/7 production since launch
• Delivered a three-tier C# supervisor/worker/client stack over RabbitMQ for through-hole automation, adapted from a sister team's pattern; live on 7+ machines across 6 lines at 2 sites
• Hardened driver connectivity with automatic TCP/IP reconnect, exponential backoff, and machine heartbeat monitoring; stateless restarts recover mid-cycle without losing machine state
• Normalized SMEMA handshakes and proprietary protocols from 30+ machines and devices across 6+ vendors into IPC-CFX, unifying inter-service messaging and Splunk/Azure telemetry
• Sole developer of the React/MUI client for a preventive maintenance platform, caching 10k+ ID-linked records in Jotai for instant local editing and multi-level revert across a machine → checklist → section → task hierarchy; used by 30%+ of manufacturing
• Directed schema and API design behind it across frontend, backend, and DBA teams: four-level many-to-many hierarchy, frequency scheduling, immutable execution records
• Migrated Flying Probe testing from WPF to a React/Redux dashboard polling machine state for board queues, barcode-miss imaging, and pass/fail across 3k+ PCBs daily
• Built a RabbitMQ/CFX developer client on personal initiative, giving the team its first practical way to test messages without a live machine; adopted by all four engineers
• Led team adoption of a sister team's Kubernetes monorepo, containerizing C# services via Docker, Jenkins, and Helm; reduced new-vendor bring-up to a normalization layer
• Mentored a junior developer building a Blazor RabbitMQ monitor with live filtering over 5k+ events and self-service driver recovery, cutting escalations to near zero

METER Group

Pullman, WA

• Leveraged React and Material UI to create ZENTRA webpages and components used by over 90% of users
• Deployed a Python weather data pipeline on AWS Lambda and DynamoDB via Serverless Framework for soil moisture predictions
• Optimized query scripts for Django models, handling over 150k objects, to assess customer habits

Claude, TypeScript, Next.js, Tailwind CSS, Vercel

• Architected a Pokémon VGC scouting web app that surfaces opponents' most common builds, runs damage calculations, and compares speed tiers to flag matchup threats against the user's team
• Kept lookups fast with a tuned caching layer over Pikalytics and Smogon data behind Next.js route handlers, auto-discovering the latest published stats period at runtime
• Eliminated redundant API calls and race conditions from out-of-order responses via a client-side caching and request-deduplication layer with optimistic loading placeholders and stale-write guards

TypeScript, Next.js, Radix UI

• Built a Pokémon team advisor web app to create type-balanced teams across all mainline games
• Designed a weighted scoring algorithm that rates each type against current team coverage, guiding users to the most type-comprehensive team