Serge Kutin

I turn ambiguous, high-stakes ideas into distributed systems that ship — and keep running in production.

For 15+ years I've owned complex software end-to-end: from the first architecture diagram on a blank page to the fault-tolerant, high-load system serving real users. I'm the person you bring in when the problem is big, the requirements are fuzzy, and someone has to be accountable for it actually working.

What sets me apart is how much I ship, and how fast. I pair deep architectural ownership — decomposition, precise specs, hard technical decisions — with an AI-assisted delivery workflow: AI agents implement features and tests under my continuous review. It's a genuine force multiplier — I run 10+ production systems in parallel, at a fraction of a traditional team's cost. Architect-grade judgment with the velocity of a small team.

Enterprise systems I've designed and delivered:

- A company-wide Enterprise Service Bus — Kafka-backed microservices on Kubernetes, with GitLab CI/CD and Prometheus/Grafana observability — plus a real-time gateway synchronizing data between an external government education platform and internal systems.

- A fault-tolerant license distribution and validation platform with centralized access control.

Products I've built end-to-end on my own — proof of the same architecture-to-production ownership across the full stack:

- A self-hosted, multi-platform messenger: Elixir/Phoenix backend (WebSocket channels, single-writer ordering) with native iOS (Swift/SwiftUI) and Android (Kotlin/Compose) clients and PostgreSQL.

- Censorship-resistant VPN infrastructure over an MQTT broker mesh, with end-to-end encryption and automatic multi-exit failover.

- Voice AI services combining GSM telephony, Whisper/VAD/STT, and LLM toolchains, orchestrated through n8n.

- An AI audio SaaS for music stem separation (Demucs) with a REST API, Telegram bot, and iOS/macOS/Android apps.

- An offline-first wishlist PWA with paired iOS and Android clients.

- Custom MCP servers that let AI agents operate real infrastructure.

I work best where the stakes are real: distributed systems, event-driven architecture, real-time messaging, and reliability under load. I ship production-grade cloud infrastructure on AWS — EC2, Lambda, S3/CloudFront, SQS/SNS, RDS/DynamoDB, ACM/IAM — plus Kubernetes, CI/CD, observability and autoscaling, architected to scale and to survive contact with production. I care about clean decomposition and honest trade-offs.

I'm looking for remote work where I can own hard problems from concept to production and quietly out-deliver expectations. If that's what your team needs, let's talk.

—

Code: github.com/mrkutin — enterprise-service-bus · mutter-phoenix · vertex · voice-assistant · stereo-hysteria · wish-with-me-web · mcp-server-qdrant