# Probably Works

> An English-language podcast by Oksana Kovalchuk and Konstantin Semenenko about building businesses, leading teams, and how AI changes design and software.

Oksana Kovalchuk and Konstantin Semenenko explore how businesses are built, how markets move, and how people make decisions when the rules keep changing. Real experiences, healthy disagreements, and a bit of sarcasm.

## Podcast

- [All episodes](https://www.probablyworks.co/episodes/): Full videos, show notes, and chapter links.
- [EP.06 — I Thought AI Would Free Me. I Was Wrong.](https://www.probablyworks.co/episodes/does-ai-save-time-at-work/): Does AI save time at work? Oksana and Konstantin compare repeatable tasks with code that still needs engineering, and ask why faster tools can leave them working longer.
- [EP.05 — CEO? More Like Chief Everything Officer](https://www.probablyworks.co/episodes/chief-everything-officer/): Oksana tells the story of starting ANODA, splitting responsibilities with a business partner, growing the team and learning when a founder needs to set boundaries.
- [EP.04 — Are You Automating Things That Don't Matter?](https://www.probablyworks.co/episodes/the-automation-trap/): Which workflows are worth automating? Oksana and Konstantin discuss process ownership, reports nobody reads and how to tell whether automation improves a result.
- [EP.03 — Why AI Won't Replace Your Design Team (Yet)](https://www.probablyworks.co/episodes/will-ai-replace-design-teams/): Oksana and Konstantin ask what a design team should deliver when AI can produce average work, from client judgment and useful feedback to taking responsibility for a result.
- [EP.02 — The Reality of Building AI Products: It’s Not Just Prompts](https://www.probablyworks.co/episodes/the-real-cost-of-building-ai-products/): What does it cost to build an AI product beyond the prompts? The hosts discuss token budgets, running models locally and the work of turning a prototype into a product people can trust.
- [EP.01 — AI vs. Humans: Who Should Control the Code?](https://www.probablyworks.co/episodes/who-should-control-ai-generated-code/): When a coding agent changes login while fixing padding, who catches it? Oksana and Konstantin discuss scope, human oversight and accountability for generated code.
- [Shorts](https://www.probablyworks.co/shorts/): Short clips from the show.
- [Meet the hosts](https://www.probablyworks.co/hosts/): Public bios and links.

## Articles

- [All articles](https://www.probablyworks.co/articles/): Practical guides inspired by the show.
- [How to decide what to automate: a break-even checklist](https://www.probablyworks.co/articles/how-to-decide-what-to-automate/): Estimate the real payback of a workflow automation, account for exceptions and maintenance, and pick a small pilot before building the whole thing.
- [A code review checklist for AI-generated changes](https://www.probablyworks.co/articles/ai-generated-code-review-checklist/): Review an AI-generated patch for changes outside its scope, verify tests against the original bug, and decide who owns the merge.

## Optional

- [Show guide](https://www.probablyworks.co/index.md): Podcast, hosts, and FAQs in Markdown.
- [Episode RSS](https://www.probablyworks.co/rss.xml): New episodes and chapter links.
- [Article RSS](https://www.probablyworks.co/articles/rss.xml): New articles.
- [WebMCP discovery](https://www.probablyworks.co/.well-known/ard.json): Read-only browser tools.
- [Sitemap](https://www.probablyworks.co/sitemap.xml): Canonical pages.
- [Video sitemap](https://www.probablyworks.co/video-sitemap.xml): Full videos and player URLs.