Dmitry
Kaznacheev
I take the technical side of a product end to end — how it is designed, how the code gets written, and what it runs on.
Co-founder & CTO at Easy and Generation One. Ten years across backend, frontend, mobile and infrastructure — most recently a Kubernetes platform running hundreds of production tenants, built from zero in eight months.
An AI agent platform. I built the platform behind it: a control plane that provisions and reconciles a dedicated instance per customer across multiple Kubernetes clusters — lifecycle operations, ingress, backup and restore, with a cluster-side agent driving Helm releases. Tenants are created and changed on the fly, so the desired state is computed at runtime rather than declared in a repo. Off-the-shelf GitOps tooling doesn't fit that shape, so I built the provisioning layer myself. Billing is wired into it: plan-driven resource limits, and provider webhooks reconciled against actual subscription state.
Hundreds of tenants in production, built from zero in eight months.
We are building Local One — a cooperatively owned food delivery platform where the restaurants own the co-ops that run it. Mobile apps for customers, couriers and restaurants, an event-driven backend and real-time tracking. Along the way I published G1.Stripe.Maui, an open-source Stripe PaymentSheet binding for .NET MAUI on both iOS and Android.
Launching 2026, piloting in Malta.
If one of these is your situation right now, writing to me is probably worth your time.
Your development process
Your team ships, but nobody can say why it takes as long as it does. I set up how work actually moves: decisions that get written down, review that catches things, tests and gates that run without being asked, releases that are boring.
AI agents in the pipeline
You want agents writing real code, not demos, and you are not willing to let quality slide for it. I run that setup daily: agents author and review, a documented pitfalls catalogue stops them repeating mistakes, and a human still owns review and release.
Architecture and technical decisions
You are choosing between approaches and the wrong one will be expensive to undo. I have made these calls under real constraints, including the ones where the off-the-shelf answer did not fit and we had to build it ourselves.
What the product runs on
Kubernetes, deployment, observability, cost. Most recently a dedicated instance per customer with everything that implies — provisioning, upgrades, backups, billing wired to resource limits — but the shape depends on what you are building.
Building the thing itself
Backend, frontend, mobile. Sometimes what you need is not advice but someone who will pick up the unfamiliar part of the product and ship it, which is most of what I have done for ten years.
Technical leadership without a team yet
You are early, the technical side is on you, and you would rather it was not. Fractional CTO work: the decisions, the process and the hands, for as many days a month as it takes.
Tell me what you are building and what is in the way.
A few lines are enough to start. I reply within a day, and the first conversation is just a conversation — if I am not the right person for it, I will say so and point you at what I would do instead.
Usually this takes one of three shapes: a few days a month as a fractional CTO, a monthly retainer, or a project with a defined end. Remote, UTC+3.