Євген

Yevhen Kravchenko
Junior AI Python Developer
[відкрити контакти](див. вище в блоці «контактна інформація») | [відкрити контакти](див. вище в блоці «контактна інформація») | [відкрити контакти](див. вище в блоці «контактна інформація»)

Wrocław, Poland | Open to hybrid/on-site/remote | Open to relocation

GitHub: https://github.com/SueRix

LinkedIn: [відкрити контакти](див. вище в блоці «контактна інформація»)

Summary
Python developer transitioning into AI automation, with 2+ years of hands-on experience across
university, bootcamp and personal projects, including 8 months in a dual-study program at a large IT
services company. Built LLM features end to end: multimodal image-to-data extraction with the Gemini
API, embedding-based search, and Celery/Redis queues that keep third-party AI calls from breaking
the app when a vendor is slow or down. Based in Wrocław, native Ukrainian, available full-time.

Ukrainian – Native | English – B2 | Polish – A1

Skills
AI & LLM Automation & Integrations
Gemini API (multimodal, image + text), LLM, RAG, REST API integration, n8n workflows, Celery + Re-
AI agents, prompt engineering, embeddings and dis (async queues, rate control, retries), provider-
vector search (cosine similarity), LangChain agnostic API adapters, Aiogram bots, web scrap-

Backend ing, CSV export

Python, Django, DRF, FastAPI, SQL, Pinecone Infrastructure, Cloud & Tools
vector DB, PostgreSQL, SQLAlchemy, Alembic, Docker/Compose, GCP, AWS, Linode, Azure,
SQLite Git/GitHub, Postman, pgAdmin, PyCharm, Ubuntu

Experience 1

Project name The Recipe Manager - personal project, 06/2025 – Present

Live demo https://50-116-58-232.sslip.io/home/

GitHub Repo https://github.com/SueRix/Favourite-Tastebook

Description Recipe recommendation and history-tracking system based on ingredient analysis from
text and images.

My contribution – Architected the Django backend on a hybrid DDD + Clean Architecture split (views
→ use cases → domain → infrastructure: AI adapters, ORM selectors, repositories,
Celery tasks), covered by 220 pytest tests.
– Built an LLM cooking agent (self-hosted n8n + Gemini) over 8 server-side tool
endpoints, and hardened that surface against prompt injection and IDOR — the
acting user travels in a signed, TTL-bounded context token rather than a tool
argument, behind fail-closed service auth and Redis rate limiting; user AI settings
are enforced in Django, not in the prompt.

Built semantic recipe search on Pinecone embeddings (orchestrated through n8n),
preserving vector ranking across Postgres hydration and calibrating similarity into a
comparable match score; a separate cosine-similarity engine ranks recipes against
the user's taste vector.
– Integrated Gemini 2.5 Flash for schema-constrained multimodal ingredient
recognition from photos, run asynchronously on Celery + Redis.
– Containerized the stack with Docker Compose (web, Postgres, Redis, Celery, n8n)
and deployed it on GCP.

Stack Python, Django, PostgreSQL, HTMX, Gemini API, Pinecone, n8n, Celery, Redis,
Docker, pytest, GCP, GitHub

Experience 2

Project name Interior design generation website - OnPoint Bootcamp, team of 16, 03/2025 –
06/2025

Description Web app for generating or improving room interiors using AI image generation.

My contribution – Built configurable request logic for the AI image-generation vendor, supporting both
custom prompts and predefined variations.
– Designed a Celery/Redis queue system to prevent API spam and protect against
vendor downtime.
– When the original AI vendor went offline mid-development, the vendor-agnostic
design allowed the team to switch to Flux 1.1 Pro with minimal delay - work praised
by bootcamp mentors.

Stack Python, Django, Celery, Redis, REST API, Git, Docker

Experience 3

Project name Pet Care Management - KhPI cooperative project, team of 5, 03/2023 – 12/2023

Description App for tracking pets' vaccinations, treatments and medical schedules.

My contribution – Built a custom Django authentication system.
– Designed a FastAPI API Gateway connecting two services built in different stacks:
C# (Entity Framework) and Python (Django).
– Deployed the project to AWS and GCP using Docker / Docker Compose;
independently resolved a cross-language integration issue.

Stack Django, DRF, FastAPI, SQLite, Docker, AWS, GCP

Other academic / cooperative projects:
- Financial Cost Management - budgeting system with Django, SQLAlchemy and Alembic migrations
(09/2024–12/2024).
- Telegram Cloth Shop - Aiogram e-commerce bot with web scraping and CSV export (03/2025–06/2025).

Bachelor's Degree | Kharkiv Polytechnic Institute (KHPI)
Education
Kharkiv, Ukraine | 09/22 – 06/2026

I hereby consent to the processing of my personal data for the purposes of the current recruitment process, in accordance with the General
Data Protection Regulation (EU) 2016/679 (GDPR).