Ілля Ветеран

ILLIA BALTIN
Trading Operations Engineer | Python Automation | Crypto PnL / QA
Ukraine · Remote / Part-time · Contract
[открыть контакты](см. выше в блоке «контактная информация») · [открыть контакты](см. выше в блоке «контактная информация») · [открыть контакты](см. выше в блоке «контактная информация»)

TARGET ROLES

Trading Operations Engineer Crypto PnL Reconciliation Bot Monitoring & Risk Ops

QA Automation Engineer Python Data Pipeline Specialist AI / Workflow Automation

PROFESSIONAL SUMMARY

QA and Python Automation Specialist with 5+ years in fintech, enterprise, and crypto operations. Expertise spans manual QA,
REST API validation, SQL-based data analysis, and end-to-end automation scripting. Currently focused on trading operations
automation: Hummingbot/Binance workflow automation, PnL reconciliation, config-level performance analytics, inventory risk
control, and decision-pipeline engineering. Proven ability to turn operational chaos into reliable, auditable systems — exactly the
discipline that separates profitable algo-trading from expensive guesswork.

PORTFOLIO PROJECT — CRYPTO OS

Crypto OS — Production analytics, risk-control and PnL reporting layer for Hummingbot Pure Market Making on
Binance Spot (Python · SQLite · Docker · Git)
• Built a deterministic Python/SQLite toolkit that reconciles Binance account trades into an accounting database with realized PnL,
mark-to-market true PnL (live Binance price fetch), and unrealized inventory valuation — eliminating a critical stale-price bug that
overstated true PnL by $29 on a $700 book.
• Designed a config lifecycle (WATCH → CANDIDATE → REFERENCE_READY) and multi-signal decision engine (market regime +
session adjuster + execution quality + spread-edge monitor + health score), producing a single FINAL_DECISION with confidence %
per pipeline run.
• Implemented HODL benchmark: every config is evaluated against passive SOL holding — ensuring active market-making genuinely
adds value beyond simple long exposure.
• Built alpha attribution engine separating realized trading alpha from market-direction PnL, revealing that v69's apparent $29 true PnL
was 52% market exposure — a critical insight for config promotion.
• Engineered session-aware spread adjuster (ASIA / EUROPE / US / LATE sessions), detecting that US session (14–20 UTC)
generated 100% of BUY_DRIFT inventory accumulation.
• Established data-integrity guardrails: config resolver, fill-drought detection, runtime YAML vs DB alignment check, config snapshot
backfill, and daily SQLite backup pipeline.
• Full audit trail in SQLite: fills · config_tests · config_profit_registry · config_lifecycle · kill_switch_log · execution_quality_decisions ·
market_regime_decisions.
• Prometheus/Grafana monitoring layer and Telegram read-only alert channel in roadmap; tagged production release
profit-pipeline-v1 on GitHub.

65+ fills 15 scripts +$8.82 profit-pipeline-v1

tracked & reconciled in prod pipeline realized alpha (v69) GitHub release

WORK EXPERIENCE

Operations & Financial Systems Specialist Nov 2024 – Present
Military Administration
• Automate taxation, billing, financial documentation, and reporting workflows using Python, reducing manual errors.
• Built an automated document submission and approval workflow for Treasury processes — cut processing time by ~40%.
• Developed multi-user automated login workflow for Cisco-based enterprise systems, improving team efficiency.
• Data extraction, validation, and operational analysis using Oracle SQL Developer; API checks with Postman.

Functional QA Engineer Sep 2022 – Nov 2024
Amdocs
 • Led end-to-end testing of web applications in enterprise fintech environments.
• Designed test strategies, plans, coverage matrices, checklists, and bug reports for complex backend systems.
• REST API testing with Postman and Swagger; traffic analysis with DevTools for frontend/backend issues.
• Collaborated with DevOps engineers; supported CI/CD quality workflows using Jenkins and Docker.

QA Engineer Aug 2021 – Sep 2022
Intellias
• Full-cycle testing of web and mobile fintech apps (iOS, Android): functional, regression, usability, compatibility.
• Python-based automation scripts reduced repetitive regression testing effort by ~30%.
• Mobile debugging using Charles Proxy, Xcode, Android Studio; defect trend analysis for release planning.

TECHNICAL SKILLS

Trading & Crypto Ops Python & Automation
Binance Spot · Hummingbot · Pure Market Making · Realized PnL Python · SQLite · Pandas · ReportLab · Automation scripts ·
· True PnL · Mark-to-Market · HODL Benchmark · Inventory Risk · Selenium WebDriver · API scripting · CI/CD workflows · Docker ·
Config Lifecycle · Spread Edge · Execution Quality · Market Git
Regime Detection

QA & Testing Data & Databases
Manual QA · Functional Testing · Regression · API Testing · SQL · PostgreSQL · MySQL · MS SQL Server · Oracle SQL
Postman · Swagger · REST · JSON · Test strategy · Bug reporting Developer · MongoDB · Amazon DynamoDB · SQLite · Data
pipelines · Validation

Monitoring & Infra Methodology
Grafana · Prometheus · Jenkins · Docker · DevTools · Charles Agile · Scrum · Kanban · SDLC · STLC · CI/CD · QA process
Proxy · Proxyman · Log analysis · Pipeline troubleshooting improvement · Operational documentation · Risk-based testing

KEY ACHIEVEMENTS

• Fixed a $29 stale-price bug in mark-to-market PnL calculation — restored true_pnl from +$29.77 to real +$0.49, enabling correct
capital-scaling decisions.
• Designed a multi-signal decision engine (15 scripts, 12 SQLite tables) producing a single auditable trading decision per pipeline run.
• Separated realized trading alpha from market-direction PnL — revealed that 52% of v69 true PnL was inventory appreciation, not
active edge.
• Reduced regression testing effort by ~30% using Python automation scripts (Intellias).
• Accelerated Treasury document processing by ~40% through automated submission workflow (Military Administration).
• Built HODL benchmark and HODL-vs-PMM comparison — first honest measure of whether market making adds value.

Education Languages

Finance — Lviv Polytechnic National University, 2000–2008 Ukrainian — Native
Java Complex — Okten Web University, 2018 English — Upper-Intermediate — technical documentation,
Software Testing — QATestLab + Lviv IT School, online meetings, written communication

Positioning: I combine QA discipline, financial accuracy, Python automation, and real crypto trading operations experience to build
reliable systems for PnL reconciliation, risk monitoring, and operational control — where software quality, financial data, and
automation intersect.