Staff Software Development Engineer in Test
I build test frameworks, CI pipelines, and release tooling that help teams find problems earlier.
Across nine years, I’ve built test infrastructure for enterprise SaaS, consumer products used by 250M+ people, mobile and backend platforms, hardware integrations, and regulated medical-device software. My current engineering projects apply that same reliability mindset to LLM applications and agentic systems—including the AI twin below.
This is what a release pipeline looks like in practice: unit, API, web and mobile tests, a performance gate, and a report a release manager can actually use.
// Illustrative simulation. Tools and stages reflect verified experience; run-specific IDs, counts, versions, and timings are fictional.
It answers from a verified profile of my experience. One model drafts each answer and a second model reviews it before it reaches you, rewriting anything it rejects. A review gate from a pipeline, applied to an agent.
An AI twin, not me; it says so when it does not know. Suggested questions use answers reviewed ahead of time. For anything that matters, email me.
Measured outcomes from production programs, not estimates.
Enterprise SaaS, consumer software at scale, and regulated medical devices. The risks were different in each environment, so the testing had to be too.
Libre glucose-monitoring ecosystem: mobile SDKs, backend APIs and hardware integrations under regulated medical-device validation. Python and Java automation, Robot Framework, Jenkins and Docker CI, cross-team automation standards and mentoring. ~70% manual regression reduction, ~30% post-release defect reduction, 95%+ CI pass rate.
UI, API, web and mobile automation for a product used by 250M+ users: Java and Selenium framework development, Appium mobile automation, and CI/CD integration across Android, iOS and web. 50% automation-coverage increase and 60% regression-cycle reduction.
Enterprise SaaS UI and API automation. Built Selenium Grid on Docker and Jenkins, API automation with REST Assured and OkHttp, deployment-pipeline automation and cross-browser testing. Coverage from ~70% to 90%; execution time from 6 hours to under 2 hours.
Web smoke, regression and acceptance automation, with cross-browser and cross-platform testing and testing support across the full SDLC.
// at work, interactive
// hands-on project work
Six principles that shape how I build reliable test systems and improve release quality.
It has users: the engineers who read its output at 2am. If the failure message does not tell them what broke and where, the test failed even when it passed.
Auto-retry hides the signal and teaches the team to ignore red. Flakes get triaged like defects, with an owner and a root cause.
A six-hour suite gets skipped before a deadline. Cutting one to under two hours at insightsoftware changed how often it ran, which is what actually caught defects.
The goal is a framework other teams adopt without me in the room. Cross-team automation standards outlive any suite I personally write.
Medical-device validation means traceability, evidence and audit trails, not just a green checkmark. It is a habit I keep even where it is not required.
At work I use Claude and Codex interactively for test-case generation, failure analysis, root-cause analysis, and code review, and review the output before it becomes part of the test system. In my LLM and agent projects the same rule becomes architecture: structured outputs, guardrails and review gates. The twin above runs every draft past a reviewer model.
Most of my work sits behind an NDA. This one does not.
A Python/pytest API test framework against a live public API: JSON Schema validation on every response, plus a Locust performance gate that fails the build on p95 latency or error rate. Runs on every push in GitHub Actions. Small on purpose — it shows how I structure a suite, not how many files I can produce.
Everything that usually takes three emails to establish.
Santa Clara, CA. Authorized to work in the US. No sponsorship needed, now or later.
Staff or Senior SDET, test infrastructure, developer automation and productivity, quality-engineering architecture, and AI quality engineering, including reliability work for AI-enabled products. Roles where the framework and CI pipeline are part of the job rather than an afterthought. Python or Java shops both work.
Both are primary. I use Python and Java across automation frameworks, API testing, CI tooling, mobile testing, and integration work, depending on the system and existing stack. My experience includes pytest and Robot Framework in Python, and Selenium, TestNG, JUnit, REST Assured, and Appium in Java. I also work with JavaScript, SQL, and Bash.
Frameworks, internal tooling, CI infrastructure, and API/SDK integrations alongside test suites. I’ve also contributed C++ unit tests for firmware-adjacent components and limited Swift and Kotlin work on an internal integration framework.
A large share. At Evernote, Appium automation with the XCUITest and UIAutomator2 drivers and CI/CD integration across Android, iOS and web releases. At Abbott, automation across the Libre ecosystem's mobile SDKs, backend APIs and hardware integrations.
Professionally, I use Claude and Codex interactively for test-case generation, failure analysis, root-cause analysis, and code review, with human review of the results. My hands-on AI engineering projects include LLM applications, agent orchestration, tool calling, structured outputs, guardrails, MCP, and multi-agent workflows; the twin on this page is one example.
It means building automation and release evidence under rigorous requirements for validation, controlled change, and traceability from requirement to test result. I apply that discipline across mobile SDKs, backend APIs, and hardware integrations in a regulated medical-device environment.
Email works best. Include the company and role so I can respond with the most relevant information.
If you’re building a team around test automation, CI, release quality, developer tooling, or AI reliability, I’d be glad to talk.