Arkadiusz Kulpa

My path into AI wasn't a straight line — it started with QA, moved through test automation, and landed in machine learning when I joined Equiniti as their first ML & AI Engineer. That blank-page brief turned into something bigger than I expected: building production AI systems and, in parallel, shaping how an enterprise actually adopts AI at scale.

The technical work spans GenAI pipelines, RAG chatbots, fraud detection models, LLM endpoint deployment, and an ongoing multi-model agentic search capability. The strategic work spans AI governance, risk and compliance frameworks, stakeholder engagement up to C-suite, and an AI Implementation Roadmap that's now the backbone of EQ's approach to rolling out AI responsibly across the business.

What I find most interesting is the gap between those two things — the point where a good technical idea meets organisational reality. That's where I do my best work.

Away from EQ, I'm building MyAI4 — a multi-agent GenAI platform I'm developing on AWS as a proper production-grade project. It runs on the Anthropic API, LangGraph, Pinecone vector stores, and a custom MLflow evaluation server I built from scratch on EC2, RDS, and S3. The frontend is a React + Tailwind SPA deployed via CloudFront. It's fully CI/CD ready, and it's mine — built because I wanted a real environment to push ideas further than a work context allows.