Антон

Anton Shkrebela
Full-Stack Developer — AI / Agentic Systems
📍 Kharkiv, Ukraine — currently Gdańsk, Poland 📞 [open contact info](look above in the "contact info" section) · [open contact info](look above in the "contact info" section)
📧 [open contact info](look above in the "contact info" section) 💻 github.com/Anton7228sk 🔗 [open contact info](look above in the "contact info" section)

SUMMARY

Computer Science engineering student at Gdańsk University of Technology, gaining hands-on experience as a full-
stack developer on a real product team (NestJS / Next.js, React, PostgreSQL). Current focus is agentic AI — building
and integrating LLM-powered agents (OpenAI, Anthropic) into production features. I work extensively with AI coding
assistants (Claude Code) to write, debug and ship code quickly, and I'm comfortable reviewing and testing what they
produce. Academic background in machine learning and computer vision (Python, TensorFlow, OpenCV). Selected for
a summer Erasmus+ research traineeship at INESC-ID (Lisbon).

E XPERIENCE

Full-Stack Developer · Eventyr (Member of Sigma Software Group) Mar 2026 – Present
AI platform for recruiting automation (HireAll) & LinkedIn account onboarding (AgentVerse)
Develop full-stack features across two products in a small team, working ticket-to-deploy: data model →
API/services → React UI → shipped to staging/production.
Agentic AI / AI-core services — build and integrate LLM agent modules through dedicated microservices
(Express + Vercel AI SDK) using OpenAI and Anthropic models, with prompt engineering, structured (Zod-
validated) outputs, human-in-the-loop escalation, and Langfuse observability.
AgentVerse — CRM/pipeline for onboarding LinkedIn accounts: contributed to a stage-machine driven prospect
pipeline, a real-time inbox (Server-Sent Events + Redis pub/sub), Telegram bot & userbot messaging (grammy,
GramJS), and third-party integrations (Unipile, Proxyline, AdsPower). Next.js 16, React 19, TypeScript, PostgreSQL,
Prisma, BullMQ/Redis, Docker
HireAll — worked on backend services (NestJS + TypeORM monorepo) and React UI under strict architecture
rules (single-responsibility services, HTTP-only controllers, transactional writes) and a CI-gated, AI-reviewed PR
workflow.
Work in a strict-TypeScript culture (no any, explicit types), write unit/E2E tests (Jest, Vitest), and use Linear/JIRA
with a sprint-branch Git workflow, Docker & DigitalOcean deployment.

RESE ARCH TRAINEESHIP

Erasmus+ Summer Traineeship · INESC-ID, Lisbon (Portugal) INCOMING Jul – Sep 2026
Machine Learning & Knowledge Discovery Group · supervisor: Prof. Arlindo Oliveira
Selected (team of 4 undergraduates) to design and build an AI-enhanced web platform for the research group —
a searchable publications/projects portal with semantic search and analytics.
Scope includes LLM-based research summarization pipelines, vector/semantic retrieval, integration with
bibliographic APIs, and AI-drafted member/alumni profiles. React / Next.js · Python backend · LLM APIs · vector search

SELECTED PROJECTS

Dog Emotion Classification (DogFACS) Sep 2025 – Present
Team of 4 building an automated emotion-classification system that analyses dog facial Action Units (DogFACS
methodology): keypoint detection for facial landmarks and a processing pipeline for 2,500+ video samples.
Python · TensorFlow · OpenCV · Keypoint Detection · Computer Vision

License Plate Recognition System Feb – Jun 2024
Automated license-plate detection and recognition (AI course project): computer-vision character segmentation and
optical character recognition.
Python · OpenCV · TensorFlow · Image Processing · OCR

EDUCATION

Gdańsk University of Technology (Politechnika Gdańska) Expected 2026
Bachelor of Engineering in Computer Science (Informatyka)
Focus on Artificial Intelligence, Machine Learning and Computer Vision. Relevant coursework: Artificial Intelligence,
Data Structures & Algorithms, Databases, Software Engineering. Currently completing engineering thesis.

TECHNICAL SKILLS

Programming Python, TypeScript / JavaScript, C++, Java, SQL
Web development React, Next.js, NestJS, Node.js, Flask
AI / ML LLM APIs (OpenAI, Anthropic), prompt engineering; TensorFlow, PyTorch, OpenCV
(coursework)
Databases & tools PostgreSQL, Redis, Docker, Git, Linux
AI-assisted development Confident with AI coding agents (Claude Code) — using them daily to build, debug and ship
real features; writing effective prompts and reviewing/testing AI-generated code before it
goes live

L ANGUAGES

Ukrainian Native
Russian Native
Polish Fluent
English B2 (Upper-Intermediate)