Evgenii Subbotin
QA Architect · SDET Lead · SAFe RTE · AI-First Engineering Leader
Who I Am
I’m an engineering leader who spent 20 years making software delivery measurably better: fewer defects in production, faster feedback loops, stronger teams, and systems that work the way they’re supposed to. I’ve done that as a hands-on SDET, as a QA architect designing test strategy for enterprise platforms, as a SAFe Release Train Engineer coordinating delivery across 8 teams and 60+ engineers, and as a de-facto Engineering Manager running the full people lifecycle. None of those roles fully describes what I do — all of them together come closer.
My default is to automate the repeatable and measure the outcome. A regression suite that takes a day to run is an engineering problem. A sprint report assembled by hand is a workflow defect. I apply the same evidence-based thinking to AI tools that I apply to test frameworks: form a hypothesis, run the experiment, log the result. My AI Engineering Log — 24 tasks per sprint with outcome per task, acceptance rate, time saved — exists because I want to know what AI actually delivers, not what its demos promise.
Right now I’m building a public portfolio that demonstrates what I can do, not just what I’ve done. Five automation stacks against the same target. Seven performance tools across five ecosystems. A production REST API built and tested solo in three days. An AI-augmented booking platform shipped in five. Every repo is public, every CI pipeline is green, every engineering decision has a documented rationale. The portfolio is the proof of work — and it’s not finished yet.
How I Compare
The radar below compares my competency profile against strong senior specialists in eight software engineering roles — not the average practitioner, but the 75th percentile: the top quarter of people who’ve spent 10 years in that specific lane.
The scale is deliberately honest: 100 means world-class, top 1% globally (the people writing the textbooks and giving the keynotes). 85–94 means principal or staff-level depth. 75–84 means strong senior. I score myself at 75–84 on my strongest axes — which is where 20 years of deliberate practice across multiple domains lands you if you’re being truthful about it.
What the diagram shows is not that I’m better than specialists at their own specialisation (I’m not — and that’s visible). It shows the profile that results from building deep expertise in QA architecture and automation, then expanding into delivery leadership, AI engineering, and requirements. Select any role to see where the profiles converge and diverge.
How I Got Here
A career that started with manual test cases in 2004 and ended up building AI-augmented systems, LLM evaluation pipelines, and multi-ART release trains — not by following a plan, but by following the interesting problems.
If You’ve Read This Far
You’ve seen the profile and the path. The next logical step is the work itself — real projects, live CI, documented decisions, and findings that weren’t hidden.