Looking for a job

Roman S'K

@romansk

Message

I build production AI/LLM systems, RAG workflows, and full-stack products from ambiguous requirements to launch.

Ukraine

Message

What I'm looking for

I'm looking to build production AI/LLM-powered product features end to end, turning ambiguous requirements into reliable systems across backend and frontend.

I've built production-style AI/LLM systems, including a RAG system from scratch and an agent that investigates incidents and generates reports autonomously.

For Corridor Watch, I built a Python/FastAPI detection service that correlated telemetry from three simulated sources through Apache Kafka, indexed data in Elasticsearch, and surfaced it through Kibana dashboards.

As the sole developer for an AI accounting platform, I designed and built the NestJS, Prisma, PostgreSQL, and React product from requirements through production, including Redis/BullMQ processing and OpenAI-powered document workflows.

Across five commercial mobile domains, I've delivered end-to-end features in Flutter and React Native while supporting REST APIs, Firebase/Supabase backends, and CI/CD to keep releases moving.

Experience

Work history, roles, and key accomplishments

Flutter / React Native Developer

Product Company

Jan 2023 - Jan 2026 (3 years)

Delivered production features across 5 commercial domains (dating, social networking, logistics, eCommerce, gaming), owning features end to end from technical breakdown to release. Worked across the stack when needed to unblock releases.

Flutter React Native REST APIs Firebase Supabase CI CD

Rail-Line Monitoring PoC

Corridor Watch

Jan 2026 - Present (8 months)

Built a detection service in Python/FastAPI/asyncio that correlates events from simulated telemetry sources streamed through Apache Kafka and indexed into Elasticsearch. Added an LLM agent that investigates flagged incidents and autonomously generates incident reports.

Python fastAPI AsyncIO Kafka Elasticsearch Kibana LLM Agents OpenAI API

AI Engineering — Applied RAG & LLM Systems

Self-directed

Jan 2026 - Present (8 months)

Built a working RAG system from scratch including tokenization, embeddings, vector search, and OpenAI API integration. Applied this to production-style projects such as an LLM agent for incident investigation and real-time game logic.

RAG OpenAI API Embeddings Vector Search Agent Design PyTorch

AI Game Director — Backend Prototype

Independent project

Jan 2026 - Present (8 months)

Built a FastAPI backend with PostgreSQL, Redis, and WebSockets, integrating the OpenAI API to drive dynamic server-side logic in real time under latency constraints.

fastAPI PostgreSQL Redis WebSockets OpenAI API

Full-Stack Developer

AI Accounting Platform

Jan 2025 - Present (1 year 8 months)

Designed and built the product's backend (NestJS, Prisma, PostgreSQL) and frontend (React 19) from scratch as the sole developer, owning the system from requirements to production. Added asynchronous job processing and integrated the OpenAI API for document understanding and generation.

NestJS Prisma PostgreSQL React Redis BullMQ OpenAI API Document Processing