Hi, I'm Andrii.
I help teams get AI working in production — not just in the demo.
Agents, RAG and MCP integrations, built into the products you already have on AWS — on infrastructure that's secure, reliable and cost-efficient.
The hard part
Most teams can build an AI prototype in a week. Getting it into production — secure, reliable, and without surprise AWS bills — is where things get hard. That's the part I enjoy most.
What I can help with
From your first agent to the infrastructure underneath it.
01
AI in production
Agents, RAG on your internal knowledge, and MCP servers that connect your product to AI tools.
02
Architecture & scaling
Serverless with Lambda, Step Functions and EventBridge; containers on ECS Fargate; event-driven design.
03
Security
Least-privilege IAM, safe tool and data access for AI agents, CI/CD with GitHub Actions and OIDC.
04
Cost optimization
Real savings on your AWS bill, without trading away reliability.
Recent work
Shipped, running, and paying for itself.
3 weeks → 1 day
Automated QA pipeline
A manual QA process replaced with an automated one on ECS Fargate, S3 and EventBridge — built for under $100 in AWS costs, so the QA team could get back to their core work.
SaaS platform · Market research company
MCP servers in production
Built and deployed MCP servers that let AI assistants work directly with two companies' products and data.
SaaS product
Agentic search
Added an agent-driven search experience to an existing product, on top of its current AWS stack.
Enterprise
RAG on internal knowledge
Retrieval-augmented search over an internal knowledge base, using managed AWS services.
How I work
Easy to start, clear all the way through.
Every engagement starts with a free intro call — no clock running, no obligation. From there, pick the shape that fits.
Good first step
Assessment
A focused review of your AWS setup or AI idea. You get a written plan: what to fix, what to build, and in what order.
AI features
Build project
A defined deliverable — an MCP server, a RAG pipeline, an agent feature — with scope and timeline agreed upfront.
Infrastructure
DevOps & cloud engineering
Hands-on infrastructure work: CI/CD with GitHub Actions, infrastructure as code, ECS and serverless setups, migrations and cost clean-ups.
Long term
Ongoing support
An architect on call: reviews, improvements, cost checks and new features as your product grows.
Fixed price or hourly — whichever works better for you and the project.
What working together looks like
01
Intro call
Free, with no clock running. Tell me what you're working on — I'll ask the awkward questions.
02
Proposal
Scope, approach and timeline in writing, priced fixed or hourly — your choice.
03
Build in small steps
Working pieces early, regular demos, no black box.
04
Handover
Documentation and a walkthrough, so your team can run it without me.
What you can count on
- Everything as code, in your own AWS account. You own it — no lock-in.
- Security and cost considered from day one, not bolted on later.
- Plain-language updates, not jargon.
- Remote, overlapping with European and US East Coast hours.
About me
15+ years of keeping production systems running.
I'm an AWS Solutions Architect and DevOps engineer based in Gdańsk, Poland. I've spent my career building and running production systems, and today I help companies bring AI into the products they already have — without breaking what works.
I work with companies of all sizes and industries. What matters most to me is working with people who want to build things properly.
If you're tackling something on AWS, AI-related or not, I'd be glad to hear about it.