Staff Engineer · Platform, Infrastructure & Developer Tooling
Boulder Creek, CA · linkedin.com/in/tedski · github.com/tedski
Summary
Staff engineer specializing in platform, infrastructure, and developer tooling with a decade of SRE experience across large-scale distributed systems. I build things that make engineers more effective, and I treat user feedback, adoption metrics, and documentation as first-class engineering concerns measured with the same rigor as uptime. My leadership instincts were forged in the U.S. Coast Guard and as a Firefighter, where the Incident Command System shaped how I approach reliability: clear ownership, documented procedures, structured on-call culture, and coordination under pressure. I spent my recent career break deepening my practice in agentic AI development, applying a structured research-plan-implement-test methodology to keep AI-assisted work genuinely maintainable.
Experience
Career Break — Personal Development & Projects
Apr 2025 – Present · Boulder Creek, CA
- Extended a home automation system spanning Z-Wave, Zigbee, and custom ESP32 hardware, building local-first replacements for cloud-dependent devices including a garage door controller, Somfy shade driver via LoRa, smart gate opener, and a circuit-level power monitor using current transformers. Iterated on automations based on feedback from household members and guests, with a design principle of graceful fallback to fully manual operation.
- Created TideSignal, an agentic AI application built entirely through Claude Code, applying a research-plan-implement-test methodology to explore agentic development patterns with sub-agents and structured workflows.
- Migrated a legacy shell-script-based fleet orchestration system to IaC using Ansible and Docker Compose (efnetmoto-fleet), closing years of personal tech debt.
- Applied production-grade operational practices throughout: documentation-driven development, no single points of failure, and infrastructure designed for handoff — same engineering rigor as enterprise systems, different scale.
- Built Jupyter notebooks to model solar PV and battery ROI using local weather station data and programmatic rate cards (solar-roi).
Staff Site Reliability Engineer → Staff Software Engineer
2019 – Apr 2025 · LinkedIn · Remote
Capacity Engineering (2021 – 2025)
- Tech lead for Platform Experience within Capacity Engineering, owning all user-facing tooling and feedback pipelines serving 1,000+ engineers across tens of thousands of services; contributed as a peer lead on Forecasting & Measurements, helping shape direction across sub-teams through structured user interviews and roadmap planning.
- Drove rightsizing and predictive auto-scaling adoption to 85% of services by building web and CLI interfaces shaped by user research — SREs preferred CLI, product engineers preferred web.
- Evangelized model outputs to service owners, building confidence in automated recommendations and dispelling skepticism to sustain adoption across the org.
- Trusted by directors and VPs as the go-to analyst for capacity investigations, using Spark, Trino, and Python to validate forecasts, diagnose anomalies, and independently verify performance improvement claims before organizational commitments were made.
- Designed a standardized metrics framework for Capacity Engineering, classifying signals into user, capacity, and health indicators to improve decision-making and dashboard consistency across the org; began implementing the ingestion pipeline before departing.
- Authored internal observability guidance adopted across Capacity Engineering, standardizing Grafana-compatible metrics and improving telemetry consistency across the team.
Resilience Engineering — Waterbear (2019 – 2021) · promoted to Staff 2019
- Founding engineer on the Resilience Engineering team; co-built LinkedOut, LinkedIn’s internal failure injection platform designed to shift resilience testing left and give developers the tooling to validate their own services.
- Contributed backend and UI features to LinkedOut’s web platform and maintained the Chrome extension, evangelizing the full platform — automated testing, group failure injection, and ad-hoc testing workflows — to drive adoption across engineering teams.
- Led development of LinkedOut’s automated test platform: scheduled Selenium-based runs across all failure modes at every call graph edge, with a custom image-comparison algorithm to surface visual regressions automatically.
- Served as tech lead for user research and roadmap prioritization; quarterly ambassador program embedded rotating engineers as co-designers, shifting LinkedOut’s culture from top-down chaos tooling to developer-owned resilience testing.
- Greenfielded a load balancer tuning tool using PySpark over historical Hive traffic data to recommend optimal tunable values per service, reducing manual tuning toil across infrastructure teams.
Senior Site Reliability Engineer
Jun 2016 – 2019 · LinkedIn · San Francisco Bay Area
- Bridged SRE and product engineering for the Content Org by shifting operational ownership to developers through tooling training and self-service runbooks, eliminating SRE release gatekeeping and increasing team autonomy and release velocity across the Publishing and Pulse stacks.
- Transitioned to the founding Resilience Engineering / Waterbear team in 2018, contributing to early LinkedOut platform development ahead of Staff promotion.
Senior Systems Engineer
Jul 2014 – Jun 2016 · Lucid Design Group (Atrius) · Oakland, CA
- Re-architected infrastructure for a fast-growing startup; introduced configuration management, containerization, and zero-touch deployment pipelines.
- Built and open-sourced internal automation tooling; evangelized DevOps and agile practices across engineering.
Lead Systems Administrator
Apr 2013 – Jul 2014 · eBay Inc · San Francisco Bay Area
- Led DevOps team of 10; overhauled change review process using GitHub PRs and Jira, improving deployment pipeline efficiency.
- Implemented infrastructure automation and config management via SaltStack; authored and open-sourced custom Salt modules.
Systems Administrator
Jul 2011 – Apr 2013 · Shopping.com (an eBay Company) · Brisbane, CA
- Re-architected feed generation application reducing server footprint from 100 to 12 nodes while improving reliability, resulting in a 95% reduction in escalations.
- Automated software releases and server provisioning, cutting deployment time 25% and new machine delivery time 50%.
Various Roles — Maritime Operations & Systems Administration
2000 – 2009 · U.S. Coast Guard · San Francisco Bay Area / New York
- Response missions: nine-year veteran conducting search and rescue, law enforcement, and homeland security operations in the San Francisco Bay Area — 800+ SAR missions, small team leadership, and sustained performance under high-pressure conditions.
- Prevention missions: Waterfront Facility Inspector at the Port of New York, leading a team of 12 managing security and safety audits across 47 facilities; implemented a standardized biometric identification system at 50+ sites while maintaining uninterrupted flow of $132B in annual trade.
Areas of Expertise
Site Reliability Engineering · Developer Tooling & Platform · Resilience & Chaos Engineering · Agentic AI Development · Capacity Engineering · Distributed Systems Observability