QA Automation Engineer | Backend Developer
Building test frameworks that scale and backend systems that deliver. Self-taught TypeScript, enterprise Java/Kotlin microservices, and Python automation specialist. Specializing in Selenium-to-Playwright migration and reactive architecture patterns.
Senior-engineering highlights: measurable impact, architecture, and code.
Challenge: Legacy Selenium suites with high flakiness and long CI times blocking fast delivery.
Solution: Designed a layered automation architecture: Gherkin feature layer, abstraction layer (reusable methods), and driver layer (Selenium/Playwright). Introduced deterministic fixtures and parallelizable test runners.
Outcome: Reduced CI test runtime by ~40% and decreased flakiness by ~60% via deterministic waits and stable fixtures.
Challenge: Unstructured tool logic inside LLM prompts causing maintainability and correctness issues.
Solution: Implemented an MCP server with Python + FastAPI to register, validate and expose mathematical tool endpoints to LLMs, enabling testable, versioned tool interfaces.
Outcome: Standardized tool access reduced logic duplication and improved integration test coverage across agent pipelines.
Challenge: Perimeter-trust assumptions creating security gaps and manual gating in CI.
Solution: Built a Zero Trust validation suite with policy-as-code, JWT-based service auth, and Docker microsegmentation for safe experiments.
Challenge: Demonstrating trade-offs between relational and reactive approaches for complex org hierarchies.
Solution: Multi-branch implementation: JPA/Hibernate (Postgres) vs Spring WebFlux (Reactive MongoDB) and a GraphQL branch — same domain, different paradigms to evaluate production trade-offs.
Outcome: Provides clear architecture comparisons helping engineering decisions; includes Docker Compose and integration test harnesses.
Challenge: Rapidly build type-safe APIs for reliable geo-data serving.
Solution: Implemented a TypeScript + Node API using Prisma for type-safe DB access and structured error handling.
Outcome: Demonstrates rapid adoption of TypeScript and production-grade API patterns.
Challenge: Securely match users with institutions while protecting user data.
Solution: Native Android app (Java) with Firebase integration, AES encryption for sensitive data, and HANLP for NLP-driven matching workflows.
Outcome: Shows end-to-end mobile + backend integration and security-first design decisions.
Specialized skill sets for QA Automation and Backend Development
AntiGravity: Advanced agentic development interface powered by Google DeepMind technology
Where I built my foundation
