Building with AI as the team.
Strategy, product, engineering, finance, marketing — one founder, AI as the team. Field notes from running JourneyBay, ProxyCraft, Kronos, and a one-person stack.
Flagship launches
What I’m building
Two products that show how I run AI as the team.
Kronos Agent OS: an operational layer for AI agents, open source
KAOS — a self-hosted Python runtime for long-running AI agents with five-layer memory, multi-agent coordination, built-in automations, and defense-in-depth security. An open source alternative to Hermes Agent and OpenClaw.
JourneyBay 2.0: An AI Travel Engine That Holds the Entire Trip in One Itinerary
JourneyBay 2.0 is live - an AI Travel Engine that edits your trip instead of just answering questions. A trip from one prompt, booking import from PDFs and photos, visa context from official sources, bring your own LLM key, and MCP access for Claude Desktop, Codex, and Cursor.
Featured reading
Where to start
Three pieces that capture how I think about AI-augmented building.
Context Engineering Guide: How to Structure Context for LLMs and AI Agents
What is context engineering and how to apply it: 6-layer context structure, 4 management strategies, persistence patterns, subagent isolation. Practical guide with Claude Code examples.
Claude Code: Complete Guide — Setup, MCP, Agents, Skills
AI Coding for Flutter: 7 Gotchas Every Guide Skips
Latest
Fresh from the workbench.
JourneyBay now works inside ChatGPT: plan trips in the conversation
AI News JourneyBayJourneyBay launched its ChatGPT app — search places, create trips, build day-by-day itineraries, and check visa context right in the chat. Also in other AI clients via MCP.
AI Startup: How to Launch a Product with a Neural Network in 2026
Cases & PracticeA practical playbook for AI startups: tech stack, unit economics, a 4-week MVP, and the mistakes that reliably kill them. Concrete numbers and recommendations.
Langfuse: A Step-by-Step LLM Observability Tutorial
TutorialsA complete guide to Langfuse: LLM request tracing, prompt management, evaluations, cost tracking. Self-hosted and cloud. Python examples.
Testing AI Agents: How to Know Your Agent Actually Works
AI OpsA practical guide to testing AI agents: unit/integration/e2e tests, eval frameworks (DeepEval 4.0, RAGAS, Promptfoo, Braintrust), metrics, CI/CD pipeline. With code examples.
How to Build Your Own MCP Server in Python: A Step-by-Step Tutorial
TutorialsA step-by-step guide to building an MCP server in Python: tools, resources, prompts, authentication, testing, and publishing. Complete working code.
Best MCP Servers 2026: Top 25 for Developers
Tool ReviewsA catalog of the best MCP servers: GitHub, PostgreSQL, Notion, Brave Search, Sentry, and 20 more. Categories, npm packages, setup.
By topic
Four pillars I write about most.
Engineering
Software engineering practices, architecture, testing
- Docker Multi-Stage Builds: From 1.2 GB to 89 MB Image
- CI/CD Pipeline Generator: AI Builds GitHub Actions Fast
- REST API Design with AI: Prompts and OpenAPI Spec Generation
Product
Product management, research, specs, prioritization
- Multiplying Free-to-Paid Conversion with AI Onboarding: A Framework
- JourneyBay 2.0: An AI Travel Engine That Holds the Entire Trip in One Itinerary
- PLG Playbook: Building an AI-Powered PQL Scoring Model
AI Ops
Running AI systems in production
- Testing AI Agents: How to Know Your Agent Actually Works
- Context Engineering vs RAG: When to Use Which
- AI Agents: The Complete Guide — What They Are, How to Build Them, Where to Use Them
Strategy
Business models, market sizing, competitive analysis
- The Solo Founder's CI Stack: Monitoring 15 Competitors for $0/Month with AI
- Positioning Statement Generator: A 4-Step AI Framework
- Value-Based Pricing with AI: How to Find the Right Price Without a Single Survey
About the author
Roman Belov — technical founder. 10+ years in product (Yandex, OZON, VK), now full-time on AI-augmented building. Writes weekly about engineering, product, and operations with AI as the team.
More about me →