Origin

My name is Artem Dolobanko. This is the story of how a DevOps engineer gradually moved into the world of development and products.

I’ve been working in IT for over 15 years. I started as a system administrator, then transitioned into DevOps, working with infrastructure, high-load systems, and teams of various sizes. Over time, I became involved not only in the technical side but also in building products — which led to the creation of several companies and projects I’m working on today.

Recently, I’ve been increasingly noticing that traditional approaches to source code management are starting to fall behind reality. This became especially clear with the rise of AI tools in software development.

My background combines infrastructure, development, and product thinking, which allowed me to see the problem from a broader perspective. Modern teams are not only dealing with the speed of code generation, but also with a lack of visibility: it’s often unclear how a piece of code was created, who is responsible for it, and whether it can be trusted.

At some point, I defined a simple goal for myself: to find a way to make AI-driven development more transparent, manageable, and predictable for teams.

Over the past year, AI has changed software development faster than any technology before it. And it forces us to rethink many привычні things.

How Origin Works

Origin sits between your AI coding agent and git. It’s not a new version control system — it’s an attribution layer that makes git AI-aware.
When a developer uses any AI coding agent, Origin’s git hooks fire automatically. The CLI detects which agent is running, reads the session transcript, and writes metadata into git notes. No manual tagging, no workflow changes. It just works.

Session Replay: See Every Prompt and Decision

Every AI coding session is captured end-to-end: the prompts the developer gave, the files the AI touched, the exact diff per prompt, token usage, cost, and model. When something breaks in production, you don’t just see what changed — you see why it changed.

Cost Visibility: Know Where Every Dollar Goes

When your team is running 5 different AI agents across 20 repos, API costs add up fast. Origin tracks cost per session, per model, per developer, per repo. No more surprise bills.

The New SCM Layer

Think about what GitHub did for git. Git existed. It was powerful. But it was a local tool. GitHub added the collaboration layer — pull requests, code review, issues, CI/CD — that made git useful for teams.

Origin does the same thing for AI coding. Git exists. It’s still the foundation. But it was built for a world where humans write code. Origin adds the AI governance layer — attribution, session tracking, cost visibility, policy enforcement, audit trails — that makes git useful for teams using AI agents.

This is the new era of source code management. Not because git is broken, but because the way code gets written has fundamentally changed, and our tools need to catch up.

What’s Next

We’re building Origin in public, and the roadmap is driven by what engineering teams actually need:

• Session chaining — automatically linking sessions that span across agent restarts and overnight breaks

• Multi-agent orchestration — tracking when multiple AI agents work on the same codebase simultaneously

• Real-time dashboards — live session streaming with token-by-token cost tracking

• Compliance reports — one-click SOC 2 and ISO 27001 evidence generation

Origin Solo is free forever for individual developers. No limits on repos, sessions, or agents. If you’re using AI to write code, you should know what it’s writing.

Get your free account →