Hi, I'm Nik 👋
Senior AI Systems Engineer building production AI systems end-to-end - from RAG and agent orchestration to fine-tuning, evaluation, inference, and deployment.
Focused on agentic workflows, LLM evaluation, and scalable AI infrastructure for data-intensive and financial use cases.
Some of my projects
Experience
London, UK
- Designing and building production-grade AI systems across RAG, agent orchestration, and LLM inference.
- Developing financial AI tooling including multi-agent equity research and analysis workflows.
- Implementing fine-tuning, evaluation, and deployment pipelines for domain-specific models.
Remote
- Built and scaled wellness platform with 140,000+ products sold.
- Led product, engineering, and full-stack platform development across e-commerce and operations.
Remote
- Built production systems for vertical SaaS platform serving 400+ golf clubs.
- Developed scalable React and GraphQL applications for bookings, operations, and staff workflows.
Bay Area / Remote
- Led team of 7 engineers building logistics and delivery platform.
- Scaled company from 4 to 100 employees with backend/data systems for real-time operations.
Bay Area, CA
- Built internal and customer-facing applications during scale-up from 50 to 450 employees.
Remote
- Led frontend delivery of marketplace MVP, launched in 2 months and onboarded thousands of users.
Graz, Austria
- Built and maintained e-commerce solutions with React, Redux, and GraphQL.
- Worked on Java and SAP Commerce (Hybris) implementations for enterprise clients.
Education
M.S. in Computer and Information Science
University of Ljubljana
Faculty of Computer and Information Science
2012 - 2015
Ljubljana, Slovenia
B.S. in Computer and Information Science
University of Ljubljana
Faculty of Computer and Information Science
2009 - 2012
Ljubljana, Slovenia
Stack

LangGraph
Agent orchestration and workflow routing

Langfuse
LLM observability, tracing, and analytics

LlamaIndex
RAG indexing and retrieval pipelines

Qdrant
Vector DB for semantic search

vLLM
High-throughput inference serving

FastAPI
API layer for production deployment

Docker
Containerized build and deployment

GCP Cloud Run
Production deployment for AI backends

Unsloth / HuggingFace
Fine-tuning workflows and model tooling

DeepEval
LLM evaluation and regression testing
Let's talk