I've long wanted a place to write about my work without the constraints of a portfolio format. A portfolio answers the question "what was built"; a blog answers "how" and "why". And "how" and "why" are usually the most valuable part: the decisions that didn't fit into a project description, the rakes I stepped on, and the approaches that saved clients hours and money.
What I'll write about
In short — about what I do every day as an AI Solution Architect. A few directions you'll see here regularly:
- AI and LLMs in practice. RAG search, AI agents, integrating GPT, Claude and Gemini into real products — what works in production and what stays a demo.
- Business-process automation. How to remove routine: pipelines, CRM/ERP integrations, bots and schedulers that run 24/7.
- Developer notes. Architectural decisions, tools, and short technical notes that explain "why this way".
- News from my projects. What I'm working on right now and how it turned out.
Why a practitioner's notes, not another AI article
Everyone writes about artificial intelligence today. The problem is that most of it retells announcements in generic terms. I'm more interested in what happens when a model has to be connected to a real knowledge base, withstand load, fit a budget and not break what already works.
If a process can be automated, it should be automated. But first it's worth understanding whether it's worth doing at all.
So the tone here will be applied: less theory, more specifics, numbers and conclusions you can actually use.
How to follow along
The blog is bilingual — there's an English and a Ukrainian version. New posts will appear here, in the Blog section. You can subscribe via RSS or follow updates on Telegram.
If there's a topic you'd like me to dig into, or a question about automation or AI in your business — drop me a line. Some of the future posts will grow out of exactly those questions.