Ukraine's #1 job service
Сергей
Vibecoder
- Considering positions:
- Vibecoder, Python-програміст, Back-end програміст, Fullstack-розробник, Full-stack програміст, Backend-програміст
- Age:
- 22 years
- City of residence:
- Odesa
- Ready to work:
- Odesa, Remote
Contact information
The job seeker has provided: Phone numberEmailMessenger
Name, contacts and photo are only available to registered employers. To access the candidates' personal information, log in as an employer or sign up.
You can get this candidate's contact information from https://www.work.ua/resumes/20160178/
Uploaded file
Quick view
version
This resume is posted as a file. The quick view option may be worse than the original resume.
Serhii Voitov
AI-Native Full-Stack Developer
Odesa, Ukraine | [open contact info ](look above in the "contact info" section) | [open contact info ](look above in the "contact info" section) | LinkedIn | GitHub
Open to remote work worldwide.
S U M M A RY
AI-native full-stack developer. I build with AI across the whole cycle: I use Claude Code and Gemini every day to plan,
write, test and ship, and I review and understand what I commit instead of copy-pasting. Strong engineering base
underneath: async Python (FastAPI), React + TypeScript, automated tests and CI. Around 6 months of hands-on
project experience and a Computer Engineering master's background. Looking for my first commercial role.
HOW I WORK WITH AI
I use AI on the full development cycle, but I stay the engineer who owns the result.
Main tools: Claude Code (daily driver) and Cursor; Google Gemini API for LLM features in my own apps.
Delegate and verify: I break a task into steps, prompt the AI in iterations, then run the code and check it before I commit. I
do not ship what I have not verified.
Claude Code skills I use most: brainstorming, multi-agent workflows, writing-plans (spec-first), systematic-debugging.
MCP integrations: Figma MCP for a web app design flow, n8n MCP for an automation pipeline.
Where AI fails: it invents methods, skips edge cases, and says "it works" without tests. So I keep engineering standards
(tests, CI, validation) on top of the speed.
EXPERIENCE
Full-Stack & AI Developer - Personal & Team Projects Jan 2026 - Present (~6 months)
DigiShop - full-stack store for digital goods with a complete payment cycle. FastAPI + async SQLAlchemy +
PostgreSQL; Redis for caching, rate limiting and as the message broker; Celery (worker + beat) for async order delivery
and stale-order cleanup; Stripe Checkout with signature-verified webhooks and idempotent payment processing; JWT
+ Google OAuth. React 18 + TypeScript (Vite, Zustand, TanStack Query, Tailwind). 134 tests (pytest + Vitest, TDD),
Alembic migrations, 5-service docker-compose, GitHub Actions CI. Verified end-to-end on real Stripe test payments.
Repo
Mail GenAIrator - full-stack AI web app that generates emails from a topic, tone and length. Next.js + React + TypeScript
+ Tailwind + shadcn/ui, Supabase, Gemini, deployed on Vercel. Added mock mode, rate-limit fallback, full error handling
and validation. Live: mail-genairator.vercel.app
Telegram Task Scheduler Bot - multi-user bot where an LLM (Gemini API) parses free text into structured tasks (strict
JSON, validation, retries). Async Python (aiogram 3, SQLAlchemy async, PostgreSQL), APScheduler. 86 tests (pytest,
TDD); deployed on a Linux VM as a systemd service with CI.
Data-collection pipeline - Python, Docker, PostgreSQL. Collects data from public APIs by profiles, dedupes via MD5 +
perceptual hashing, with a moderation panel and auto-publishing. Explored n8n (with MCP), then moved the logic to
Python.
LoLAnalitik (team, Backend Developer) - backend for an ML game-analysis tool. Built a TTL/LRU cache that cut
external API load 3×; nearest-position lookup with scipy cKDTree at O(log n). First team project with shared codebase
and code reviews.
E D U C AT I O N
Master's - Computer Engineering (IT) | ONTU, Odesa Sep 2025 - Present
Bachelor's - Game Development & Interactive Media (VR) | ONTU, Odesa Sep 2021 - Jun 2025
C E R T I F I C AT I O N S
Claude Code 101 - Anthropic | May 2026
Claude Code in Action - Anthropic | April 2026
TECHNICAL SKILLS
AI-native workflow: Claude Code, Cursor, Google Gemini API, prompt engineering, context management, LLM integration,
structured outputs, MCP (Figma, n8n), multi-agent task splitting
Languages: Python, JavaScript, TypeScript, C# (OOP), SQL
Backend / API: FastAPI, aiogram 3, REST API, Pydantic v2, SQLAlchemy (async), Alembic, Celery, pytest, TDD
Frontend: React 18, Next.js (App Router), TypeScript, Vite, Zustand, TanStack Query, Tailwind CSS, shadcn/ui, Vitest
Databases: PostgreSQL, Redis, Supabase, SQLite
Payments / Auth: Stripe API (Checkout, webhooks, idempotency), JWT, OAuth 2.0 (Google)
DevOps: Docker, docker-compose (multi-service), GitHub Actions (CI), Vercel, systemd, Linux, Oracle Cloud
Algorithms & Data: graph algorithms (DFS, cycle detection), spatial indexing (scipy cKDTree), TTL/LRU caching, web
scraping (Playwright)
Tools: Git, GitHub, GitLab, VS Code, Figma
LANGUAGES
Ukrainian - Native
Russian - Native
English - Intermediate (B1)
AI-Native Full-Stack Developer
Odesa, Ukraine | [
Open to remote work worldwide.
S U M M A RY
AI-native full-stack developer. I build with AI across the whole cycle: I use Claude Code and Gemini every day to plan,
write, test and ship, and I review and understand what I commit instead of copy-pasting. Strong engineering base
underneath: async Python (FastAPI), React + TypeScript, automated tests and CI. Around 6 months of hands-on
project experience and a Computer Engineering master's background. Looking for my first commercial role.
HOW I WORK WITH AI
I use AI on the full development cycle, but I stay the engineer who owns the result.
Main tools: Claude Code (daily driver) and Cursor; Google Gemini API for LLM features in my own apps.
Delegate and verify: I break a task into steps, prompt the AI in iterations, then run the code and check it before I commit. I
do not ship what I have not verified.
Claude Code skills I use most: brainstorming, multi-agent workflows, writing-plans (spec-first), systematic-debugging.
MCP integrations: Figma MCP for a web app design flow, n8n MCP for an automation pipeline.
Where AI fails: it invents methods, skips edge cases, and says "it works" without tests. So I keep engineering standards
(tests, CI, validation) on top of the speed.
EXPERIENCE
Full-Stack & AI Developer - Personal & Team Projects Jan 2026 - Present (~6 months)
DigiShop - full-stack store for digital goods with a complete payment cycle. FastAPI + async SQLAlchemy +
PostgreSQL; Redis for caching, rate limiting and as the message broker; Celery (worker + beat) for async order delivery
and stale-order cleanup; Stripe Checkout with signature-verified webhooks and idempotent payment processing; JWT
+ Google OAuth. React 18 + TypeScript (Vite, Zustand, TanStack Query, Tailwind). 134 tests (pytest + Vitest, TDD),
Alembic migrations, 5-service docker-compose, GitHub Actions CI. Verified end-to-end on real Stripe test payments.
Repo
Mail GenAIrator - full-stack AI web app that generates emails from a topic, tone and length. Next.js + React + TypeScript
+ Tailwind + shadcn/ui, Supabase, Gemini, deployed on Vercel. Added mock mode, rate-limit fallback, full error handling
and validation. Live: mail-genairator.vercel.app
Telegram Task Scheduler Bot - multi-user bot where an LLM (Gemini API) parses free text into structured tasks (strict
JSON, validation, retries). Async Python (aiogram 3, SQLAlchemy async, PostgreSQL), APScheduler. 86 tests (pytest,
TDD); deployed on a Linux VM as a systemd service with CI.
Data-collection pipeline - Python, Docker, PostgreSQL. Collects data from public APIs by profiles, dedupes via MD5 +
perceptual hashing, with a moderation panel and auto-publishing. Explored n8n (with MCP), then moved the logic to
Python.
LoLAnalitik (team, Backend Developer) - backend for an ML game-analysis tool. Built a TTL/LRU cache that cut
external API load 3×; nearest-position lookup with scipy cKDTree at O(log n). First team project with shared codebase
and code reviews.
E D U C AT I O N
Master's - Computer Engineering (IT) | ONTU, Odesa Sep 2025 - Present
Bachelor's - Game Development & Interactive Media (VR) | ONTU, Odesa Sep 2021 - Jun 2025
C E R T I F I C AT I O N S
Claude Code 101 - Anthropic | May 2026
Claude Code in Action - Anthropic | April 2026
TECHNICAL SKILLS
AI-native workflow: Claude Code, Cursor, Google Gemini API, prompt engineering, context management, LLM integration,
structured outputs, MCP (Figma, n8n), multi-agent task splitting
Languages: Python, JavaScript, TypeScript, C# (OOP), SQL
Backend / API: FastAPI, aiogram 3, REST API, Pydantic v2, SQLAlchemy (async), Alembic, Celery, pytest, TDD
Frontend: React 18, Next.js (App Router), TypeScript, Vite, Zustand, TanStack Query, Tailwind CSS, shadcn/ui, Vitest
Databases: PostgreSQL, Redis, Supabase, SQLite
Payments / Auth: Stripe API (Checkout, webhooks, idempotency), JWT, OAuth 2.0 (Google)
DevOps: Docker, docker-compose (multi-service), GitHub Actions (CI), Vercel, systemd, Linux, Oracle Cloud
Algorithms & Data: graph algorithms (DFS, cycle detection), spatial indexing (scipy cKDTree), TTL/LRU caching, web
scraping (Playwright)
Tools: Git, GitHub, GitLab, VS Code, Figma
LANGUAGES
Ukrainian - Native
Russian - Native
English - Intermediate (B1)
More resumes of this candidate
Considering positions: Fullstack-розробник, Vibecoder, Python-програміст
Odesa, Remote
Serhii Voitov AI-Native Full-Stack Developer Odesa, Ukraine | | | LinkedIn | GitHub Open to remote work worldwide. SUMMARY AI-native full-stack developer. I build with AI across the whole cycle...
Velykyi Dalnyk, Remote
СЕРГІЙ ВОЙТОВ Junior Python / C# Backend розробник Одеса, Україна • • ПРОФЕСІЙНИЙ ПРОФІЛЬ Junior Python/C# розробник з практичним досвідом створення production-рівневих backend-систем. Спеціалізація...
Similar candidates
-
Vibecoder
40000 UAH, Remote -
Lovable Website vibecoder
Remote
The candidate's resume is posted at https://www.work.ua/resumes/20160178/