Full stack & platform
Web and mobile, data, auth, payments, and the deployment story: containers, CI/CD, and observability so AI workloads behave in production.
- Next.js
- Kubernetes
- CI/CD
- Auth & Payments
Dimitar Petrov
Full StackAI Engineer
11+ years building full-stack products end to end · AI integrations, RAG, and production-grade engineering. Web, mobile, and enterprise platforms from scoped MVP to production.
Open to remote contracts with full ownership of the delivery.
Next.jsReactReact NativeExpressFastAPIPythonRAGKubernetes
Full Stack through productionapps, data, ML
From scoped MVP to production platforms · web, mobile, AI, and the DevOps path that keeps it running.
AI & ML in production
Models and features users touch: RAG, LLM APIs, evals, and guardrails, with monitoring and safe rollouts so shipping stays predictable.
- RAG
- OpenAI
- LLM APIs
- Evals
Mobile Apps
React Native for iOS and Android: store-ready builds, native UX patterns, and APIs wired to your product - whether mobile-first or alongside the web.
- React Native
- iOS & Android
- Store builds
MVP in 1 Week
From concept to working product. Validated, tested, and ready for users.
- 1 week
- Validated
- User-ready
AI-PoweredDevelopment
I build with the latest AI-native development practices, treating Claude and Cursor not as autocomplete tools, but as collaborative agents throughout the entire development lifecycle.
Cursor
Claude
MCP
Agents
Collaborative agents across the full build lifecycle.
AI Agents & Automation
I design and deploy AI agents that handle multi-step tasks autonomously, from data pipelines to automated decision-making workflows. Claude acts as an orchestrator, breaking down complex goals into executable subtasks.
- Anthropic
- LangChain
- n8n
AI-Assisted Development
My primary dev environment pairs Cursor IDE with Claude Sonnet, enabling a fast, iterative development cycle driven by natural language. I focus on architecture and product decisions while AI handles boilerplate, refactoring, and code review in real time.
- Cursor
- Claude
- Review
Prompt Engineering
Effective AI output starts with precise input. I apply structured prompting techniques: chain-of-thought, few-shot examples, role framing, and XML-tagged context to get consistent, production-ready results from frontier models.
- XML tags
- Evals
- Guardrails
MCP & Tool Integrations
I work with Model Context Protocol (MCP) to connect Claude to external tools, APIs, and data sources, enabling AI workflows that interact with real systems rather than operating in isolation.
- MCP
- APIs
- Integrations
Scope
Build
Ship
Scale
11 years ofdelivered work
21+ production projects across web, mobile, AI, and data.
Let's talk
Ready to build somethingexceptional?
Ready to discuss your project? Pick a time below.
Dimitar Petrov
Founder
Based in Sofia, Bulgaria
- Traveller
- CrossFit
- 11+ Years Dev
- Coffee Lover
- Startup Enthusiast
- Car Enthusiast