Resume

Resume

Applied AI and AI Evaluation engineer combining 10+ years of software quality, automation, API, CI/CD, and release experience with an M.S. in Data Science, AI, and Machine Learning from Boston University.

M.S. Data Science, AI & ML — Boston University

Download PDFOpen Resume PDFSee flagship AI projectContact Me
Open to full-time US remote opportunities.Primary: Applied AI · AI Evaluation · AI Platform. Quality specialty: LLM evaluation · AI quality · source grounding · data quality.EmailLinkedInGitHub

Summary

Lead software-quality engineer with 10+ years of professional experience in automation, APIs, integrations, CI/CD, release controls, and evidence-driven validation. Applies that foundation to authorized internal AI-assisted workflows at Advisor360° and independently builds agent, evaluation, provenance, and observability infrastructure through World Publishing Houses.

Target roles

Primary focus

Applied AI · AI Evaluation · AI Platform

Strongest fit: teams building production AI systems that need evaluation, source grounding, prompt governance, observability, regression testing, data quality, and safety boundaries.

Quality specialty

LLM Evaluation · AI Quality · Data Quality · Source Grounding

Professional Experience

Full resume available as PDF.

Lead QA Automation Engineer

Advisor360° · Weston, MA
Apr 2022 – Present

  • Design and maintain test strategies, automation frameworks, and system integration/regression coverage for complex fintech web, mobile, REST API, and backend workflows using Python, Appium, Docker, Swagger/OpenAPI, and CI/CD tooling.
  • Partner with Product, Engineering, and Business stakeholders to translate requirements and operational risks into acceptance criteria, test scenarios, regression coverage, and release evidence.
  • Configure and evaluate authorized internal AI-assisted meeting workflows involving prompts, extraction logic, structured-output schemas, context boundaries, and expected behavior.
  • Compare generated summaries, action items, and structured records with source transcripts and downstream requirements for completeness, factual consistency, attribution, unsupported content, schema compliance, and operational usability.
  • Investigate failures across context assembly, retrieval, permissions, tool calls, APIs, transformations, persistence, logs, and database records rather than treating every incorrect result as a model failure.
  • Convert important findings into reproducible regression scenarios and communicate severity, affected workflows, supporting evidence, and release risk to cross-functional stakeholders.

Independent Builder · Applied AI / AI Platform Engineer

World Publishing Houses — Independent Project
Current

  • Build a source-backed full-stack publishing intelligence platform using Python, FastAPI, SQLAlchemy, PostgreSQL, Next.js, and TypeScript.
  • Design controlled AI research and tool-calling workflows with source grounding, safe abstention, deterministic rights-safety validation, and human review.
  • Implement prompt governance with immutable versions, hashes, evaluation-gated promotion, execution provenance, audit history, and rollback.
  • Build AI telemetry for traces, models, prompt versions, tokens, estimated cost, latency, failures, model pricing, and workspace budgets.
  • Build an evaluation platform with golden datasets, deterministic and model evaluators, regression comparison, human review, failure taxonomy, and quality gates.
  • Develop automated data-quality audits and ingestion/review workflows for multilingual publishing data, provenance, duplicate detection, and relational integrity.
View flagship case studyWatch product demo

Lead QA Automation Engineer

Peapod Digital Labs · Quincy, MA
Oct 2017 – Apr 2022

  • Led test strategy, manual and automated quality activities, defect triage, and release validation for large-scale e-commerce web and native mobile products.
  • Built Java-based Selenium and Appium frameworks and integrated regression coverage with TestNG, Maven, Bamboo, Sauce Labs, and Selenium Grid.
  • Automated functional, integration, REST API, end-to-end, cross-browser, database, performance, and production-smoke tests.

Software QA Tester

Infomatrix Global · Boulder, CO
Mar 2015 – Oct 2017

  • Built and executed Java/Selenium automated and exploratory/manual tests across the SDLC; performed functional, regression, integration, REST API, load, and performance testing with Jenkins, ReadyAPI, BlazeMeter, and JMeter.

Technical skills

Applied AI & evaluation

AI agents, tool calling, LLM evaluation, evaluation harnesses, prompt governance and regression, golden/reference datasets, source grounding, hallucination and unsupported-claim detection, deterministic guardrails, failure-mode analysis, AI telemetry, latency/cost evaluation, and human review.

Software & quality engineering

Python, Java, TypeScript/JavaScript, FastAPI, Next.js, React, pytest, Playwright, Selenium, Appium, REST Assured, Cucumber, TestNG, JUnit; test strategy, API contract validation, CI/CD quality gates, regression, integration, E2E, accessibility, performance, and release testing.

Data & platforms

SQL, PostgreSQL, SQLAlchemy, Alembic, pandas, NumPy, scikit-learn, data profiling, schema validation, ETL/ingestion validation, anomaly detection, EDA, feature engineering, predictive-model evaluation, Docker, GitHub Actions, Jenkins, Bamboo, Bitbucket, Postman, Swagger/OpenAPI.

Engineering strengths

  • Evidence-based failure analysis and root-cause investigation.
  • Regression and release gates for deterministic and nondeterministic systems.
  • Source provenance, data integrity, and reproducible investigations.
  • Cross-functional communication across Product, Engineering, QA, and Data.

Education

Boston University
M.S. in Data Science, AI, and Machine Learning

Quincy Community College
Computer Science & Programming · 2020

Agricultural University, Mykolaiv, Ukraine
Degree in Agricultural Business · 2011

Languages

English · Ukrainian · Russian.

Portfolio proof

Start with World Publishing Houses for the full Applied AI platform story, then explore the ML model-validation work.

WPH flagship case studyML validation case studyDownload PDF