Software engineer building distributed backend systems — and increasingly focused on developer experience, platform, and infrastructure work: the tools, pipelines, and services that make other engineers faster and systems more reliable.
- 🔭 Currently: Member of Technical Staff @ Envoy (Oct 2025–present)
- 🌱 Exploring: developer experience, platform engineering, infrastructure, distributed systems, operating systems, storage
Primary owner of the Reservations service and the Maps service, powering enterprise workplace reservations and floor map view/edit operations.
- Cut floor map view API P95 latency from 3s → 500ms with service-level caching, PostgreSQL indexes for common query patterns, and parallelized scatter-gather requests across three backend services — supporting 28.5M monthly requests.
- Leading the rollout of a Next.js + MapLibre floor map viewing and editing
frontend replacing Envoy's legacy floor map viewer/editor across live, shared,
and kiosk views — reaching ~9M employees across ~87K companies.
- Improved map pan, zoom, and tap frame rate on low-end kiosk displays by ~17% (46 → 54 FPS) by cutting per-frame main-thread work: rendering selected markers in a lightweight overlay instead of re-sending all marker data, and skipping label recomputation and re-renders unless the data actually changed.
- Led a migration of Envoy's Next.js apps to versioned static asset hosting (S3 + CloudFront), so open sessions no longer break when a new version rolls out. Built shared CI workflows and rolled them out across 7 repos that deploy about 200 times a month.
- Designed and built an asynchronous Kafka-based message routing platform for reliable bidirectional communication between Dasher app chat and SMS, with ordered delivery per conversation, idempotency, and retry semantics — saving ~$1M/year in delivery costs.
- Developed production-scale load tests that simulated peak production traffic against the chat platform's golden paths, run monthly to continuously validate reliability, latency, and scalability.
- Automated customer support workflows that increased self-service adoption, reducing support costs by ~$1M/year.
- Languages: Java, Kotlin, SQL, TypeScript / JavaScript, Python, C
- Backend: Spring Boot, REST APIs, Kafka, PostgreSQL, event-driven architecture, distributed systems
- Infrastructure: Docker, ArgoCD, Git, CI/CD
- Frontend: React, Next.js
- Education: B.A. Computer Science, UC Berkeley (2019–2023) — Database Systems, Operating Systems
- 📧 wlouis23@berkeley.edu
- 💼 linkedin.com/in/williammlouis