Yaroslav

Yaroslav(Jaroslaw) Slipchuk
Senior / Lead AI & Software Systems Engineer
(AI Systems | LLM/Agentic Workflows | Distributed
Backend | Edge AI | Node.js | TypeScript | C/C++ | Python)
Contact information: Wrocław, Poland | [open contact info](look above in the "contact info" section) | [open contact info](look above in the "contact info" section)
LinkedIn: [open contact info](look above in the "contact info" section) | Portfolio: [open contact info](look above in the "contact info" section)

Executive Summary
● Senior / Lead AI & Software Systems Engineer with 17+ years of experience in
software and hardware engineering, including 6+ years building AI-powered products
and 10+ years designing production edge-cloud, real-time and distributed systems
across EU markets. Hands-on experience with LLM integrations, RAG pipelines,
agentic workflows, prompt engineering, local inference with Ollama & vLLM,
computer vision and Edge AI. Strong startup background in rapid prototyping, product
design, MVP development, and scaling products in real-world environments.
● Experienced in pre-sales, technical discovery, solution architecture, PoC design, effort
estimation, and customer workshops, translating business requirements into scalable
AI, IoT, and cloud-native solutions. Participated in pre-sales for commercial AI products,
including autonomous retail stores and AI-powered vending platforms, collaborating with
business stakeholders from concept through production delivery.
● Delivered production AI systems operating 24/7, including an autonomous retail platform
processing 140+ camera and 450+ sensor streams in real time and an AI vending platform
deployed across 100+ machines in the EU.
● Strong background in designing and owning mission-critical systems integrating AI, edge
computing, and hardware-driven data flows.
● Proven ability to lead small to mid-size engineering teams, make architecture
decisions, and deliver reliable backend platforms from concept to production.
● Domain expertise across autonomous retail, vending, hospitality, IoT, payments and
adjacent FinTech, with hands-on experience integrating software, AI, edge devices and
payment infrastructure.

Technical Skill
Skill Category Skill List
Programming Proficient in Node.js / JavaScript / TypeScript (7+), NestJS, Python
Languages
Experienced: C programming for real-time systems and embedded
applications and C++ for Arduino projects (9 years).
AI / LLM Engineering LLM integrations (OpenAI, local LLMs via Ollama, vLLM) (2+), RAG
pipelines(2+), Prompt engineering(3+), Embeddings / vector search(1+),
AI agents / workflow implementation, Context engineering(3+), Local
inference pipelines(3+), LangChain, MCP / Function & Tool calling,
Structured Outputs / Schema Validation
Databases & Brokers Experienced in NoSQL databases (MongoDB, Redis) and SQL-based
& Communication systems (PostgreSQL, SQLite), RabbitMQ (2 years), Solace / MQTT (2+
framework/protocols years), Express.js(4 years), Socket.IO(WebSocket) (6+ years)

Operating Systems Proficient in Linux (Ubuntu, Linux for Single-board Computer and
servers) (7+)
Experienced in MacOS (2 years).
Containers & Proficient in Docker (4+ years), Docker Compose (2+ years)
Orchestration Experienced in Kubernetes
Cloud & Google Cloud, Heroku, AWS (trained/certified), On-Premise & Edge
Infrastructure Infrastructure(7+), VPN, VPS.

Development Tools / Visual Studio Code (7+), GitHub Copilot (2+ year), ChatGPT (2+ years),
AI-Assisted Claude (2+ year), Cursor(2+)
Engineering
Collaboration & Agile GitLab (7+), Scrum (4 years), Jira(7+), Slack/Google Chat (6+),
Practices Stand-ups (10+ years).

I agree to the processing of personal data provided in this document for the purposes of the recruitment process in accordance with the Personal Data Protection Act of 10 May 2018 (Journal of
Laws 2018, item 1000) and the Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 (General Data Protection Regulation)
Testing Methodology Test-Driven Development (TDD)(8+ years), AI-TDD(1+ years), Unit
/ Observability testing(2+ years), Integration testing(7+ years), E2E testing(2+ years),
Prometheus/Grafana (2+)
Media Framework ffmpeg(5+), gstreamer(2+), video and audio streaming(5+ years).
CI/CD Jenkins, GitLab CI/Webhooks, automated build/test/deployment
pipelines

Language knowledge
Language Level Experiences
English Advanced 336 hours of education in American English Center , 2 weeks 🔗
of business course in Malta, 1.5 years of work experience in
Germany, 10+ years of experience in international companies,
C - level communication, Client communication, Team
communication
Polish Intermediate
Ukrainian Native
Russian Advanced

Education
Education Specialization, awards, degree Estd. Grad
institutions
Lviv Polytechnic Information communications networks(Professional in 2007 2012
National University electronics and telecommunication), Master's degree . 🔗
Сourse leader. 2nd position in Physics Olympiad 2008.
Cisco programs(CCNA).
Lviv college of Technical support of computers and communication. 2003 2007
communication Bachelor's degree.

Certification
🔗
AWS Training & Certifications: AWS Cloud Practitioner Essentials ; Generative AI with 2025
🔗
Diffusion Models ; Domain 1 Review: AWS Certified Machine Learning - Specialty ; 🔗
Software architecture in practice at Robot_dreams by Slava Agafonov (Architect at 2023
Microsoft) 🔗
Data Structure and Algorithms at Projector by Dmytro Korduban(Technical staff at 2018
OpenAI)
The certificate for completed the advanced English level 🔗 2017
Business English Certificate in Malta 🔗 2016
The Minor Academy of Sciences in Lviv. Computer and programming sciences. 2nd 2002
position in Programming Olympiad 2003. -200
3

Work experience
Company name Position Years (from-to)
Walkout Technologies Senior Software Engineer → Tech Lead → Team Lead 2018-2026
GmbH(Sterlix - Led architecture and development of production
AI-powered autonomous retail, vending and self-service
GmbH/Smoothr, Schwarz platforms.
Digits, PanOston B.V.) - Designed distributed Node.js / TypeScript / Python
backend systems, microservices and event-driven
architectures.
- Led delivery across backend, AI, Edge/IoT and
embedded components from PoC/MVP to commercial
production.
- Collaborated directly with C-level stakeholders,
customers and external architects.
- Supported production deployments across EU markets,
including 24/7 AI systems and 100+ commercial vending
installations.
Robot_dreams Lecturer for Author Course on Backend Development 2021
with Node.js 🔗
Heinzmann GmbH & IoT / Firmware developer 2014-2018
Co.KG
UNIPAY Team leader of the Firmware & Hardware 2009 - 2014
development.

2
Selected Projects | View Portfolio

AI-Powered Recruitment Platform — Candidate Matching, Risk
Analysis & Hiring Insights
Role: Founder | Senior Software Engineer — Backend / AI Systems

Designed and built an AI-powered HR / recruitment platform that integrates with ATS systems (PeopleForce), analyzes
applicant CVs at scale, and produces structured hiring insights for recruiters. The system ranks candidates against a
vacancy, calculates a matching score for each applicant, and surfaces strengths, skill gaps, experience risks, and hiring
recommendations to accelerate shortlisting.

Key Contributions:

● Designed modular multi-service architecture: ATS sync, document conversion, profile enrichment, candidate
matching, interview question generation, and report generation
● Built PeopleForce ATS integration to sync vacancies, applications, candidate profiles, and CV attachments
● Implemented PDF-to-Markdown CV conversion pipeline (Docling) with structured LLM-based profile enrichment
(location, seniority, industries, work authorization)
● Developed LLM matching workflows (Ollama multi-GPU local inference and OpenAI) with Zod-validated
structured outputs: matching score (0–100), strengths, risks, skill gaps, and hiring recommendation
● Generated recruiter-facing CSV and standalone HTML reports with ranking, filtering, risk highlighting, and
interview clarification questions
● Defined product roadmap toward usage-based SaaS (OAuth, company onboarding, ATS multi-provider support,
pay-per-analyzed-candidate billing)
● Profiled and optimized local LLM inference using tokens/sec, TTFT, memory utilization, and inference cost as
key metrics

Tech Stack:
Node.js, TypeScript, REST, PeopleForce API, Docling, Ollama, OpenAI, Zod, Prompt Engineering, Structured Outputs, Local
LLM Inference, CSV/HTML Report Generation, Vitest

AI Anti-Theft system for Self-Ordering Kiosk
Role: Team Lead | Architect | Senior Software Engineer

Designed and contributed to a real-time fraud detection subsystem for a large-scale retail software platform. The system
processes live camera streams, analyzes video data to detect potential customer fraud, and publishes detected events to
MQTT for further handling and integration.

Key Contributions:

● Collaborated with third-party architects and C-level stakeholders.
● Developed automated system and functional tests using Node.js, TypeScript, and Python
● Built CI/CD pipelines using Jenkins and GitLab webhooks
● Built a video generation system for E2E system testing
● Contributed to AI-powered video services using Python
● Built automated labeling workflows for ML model training with Python and PyTorch
● Worked with real-time video processing at edge servers
● Implemented GPU-accelerated video encoding and decoding using CUDA in Docker containers

Tech Stack:
Node.js, TypeScript, Python, Socket.IO, MQTT, Docker, Docker Compose, Kubernetes, Microservices, Event-Driven
Architecture, CUDA, FFmpeg, GStreamer, PyTorch, Jenkins, GitLab.

AI Drive-In — AI Voice Agent for Restaurants
Role: Senior AI & Backend Engineer

Built a prototype voice AI agent for restaurant drive-in ordering. The system automatically communicates with customers,
processes food orders at the checkout gate, and integrates real-time audio processing with LLM-based workflows.

Key Contributions:

● Collaborated with C-level stakeholders.
● Designed event-driven backend services for real-time voice interaction
● Integrated audio streaming pipelines using WAV chunks and FFmpeg
● Implemented Voice-to-Text and Text-to-Voice processing flows
● Integrated Ollama-based local LLM workflows
● Built AI agent and RAG-based functionality
● Worked with edge AI computing using Raspberry Pi and sensor fusion

3
Tech Stack:
Node.js, TypeScript, REST, Socket.IO, FFmpeg, Raspberry Pi ,Banana Pi BPI-M6, CUDA, Sensor Fusion, Voice-to-Text,
Text-to-Voice, Ollama, AI Agents, RAG, PM2

Unmanned Autonomous AI Store Platform
Role: Team Lead | Architect | Senior Software Engineer

Designed and led the development of a real-time AI video analytics system operating 24/7 for an autonomous retail store
deployed in production in Stuttgart.

The system processed 140+ camera streams and 450+ sensor streams in real time to support automated checkout.

Key Contributions:

● Collaborated with C-level stakeholders
● Designed solution and software architecture
● Built microservice and distributed system architecture
● Implemented event-driven communication between services
● Designed real-time video and sensor processing at edge AI servers
● Integrated computer vision and sensor fusion pipelines
● Worked with GPU-accelerated video encoding and decoding CUDA
● Led technical delivery across backend, AI, and embedded components
● Profiled and optimized edge AI workloads for latency, memory usage, throughput, and power consumption on
ARM/NPU-based hardware

Tech Stack:
Node.js, TypeScript, REST, Socket.IO, Express.js, FFmpeg, TCP/IP, Raspberry Pi, GPU Acceleration, Computer Vision,
Sensor Fusion, Edge AI, Microservices, CUDA, Python/Flask, YOLOv8 ONNX deployment on ARM/TIM-VX NPU, Web(JS,
CSS, jQuery, React).

Autonomous AI Vending Platform
Role: Tech Lead | Architect | Senior Software Engineer

Designed and developed an autonomous vending machine platform using computer vision, weight sensors, and advanced
algorithms for seamless checkout and accurate transactions.

The product was deployed across 100+ vending machines in the EU and continues to be sold commercially.

Key Contributions:

● Collaborated with C-level stakeholders
● Designed solution and software architecture
● Built microservice and distributed architecture
● Developed event-driven backend systems
● Integrated computer vision, sensor fusion, and real-time edge/cloud processing
● Implemented production monitoring using Prometheus and Grafana
● Integrated payment systems, including Worldline
● Supported product scaling from MVP to large-scale EU deployment

Tech Stack:
Node.js, TypeScript, C/C++, Redis, MongoDB, Docker, Jest, Prometheus, Grafana, Socket.IO, Express.js, FFmpeg, TCP/IP,
Arduino, ESP12E, Computer Vision, Sensor Fusion, Worldline, PM2.

4