I architect and build scalable AI-powered platforms processing thousands of daily events for enterprise clients. From leading microservices design to optimizing infrastructure costs, I transform complex business requirements into reliable, production-ready solutions.
I developed a unified agent service capable of representing itself as multiple specialized agents within a single architecture. The system acts as a centralized service that exposes distinct agent personalities dynamically, depending on the client and context. Each tenant can invoke its own agent identity, allowing flexible task execution and individualized behavior.
Instead of developing new agents from scratch, the system enables teams to focus on building tools and extensions, dramatically accelerating iteration and scalability.
Key achievements include:
Technologies used: Docker • FastAPI • OpenAI API • Redis • asyncio • Logfire • MCP • Tavily • UV • Python • Pydantic AI • Azure Bot Service • RAG • BullMQ
As a technical lead, I redefined and scaled an AI platform for agents that connects multiple client-side applications and services through a unified integration layer. The platform aggregates messages from connected apps, identifies users, and routes tasks smoothly between internal services such as agent engines.
My focus was on improving scalability, simplifying architecture, and closing security and performance gaps while ensuring the system could support hundreds of concurrent users.
Key achievements include:
Technologies used: Node.js • Python • Fastify • Express • FastAPI • Docker • Redis • PostgreSQL • Azure Bot Service • Azure Bot Framework • Azure Blob Storage • GCP Cloud Run • GCP Gateway • GCP IAM • Logfire • LangChain • LangGraph • LangFlow • Tavily • OpenAI API • Pydantic AI • Qdrant • MCP • A2A
I delivered a self-hosted AI platform designed for enterprise companies that require full data isolation and on-premise deployment. The system allows clients to run private LLMs and pipelines entirely within their secure infrastructure, ensuring zero data retention, complete isolation, and reliable performance comparable to cloud environments.
Key achievements include:
Technologies used: Docker • Kubernetes • PostgreSQL • Redis • Fastify • Python • SharePoint • LangGraph • RAG • Pydantic AI • Self-hosted Mistral • Azure • Custom UI
I developed an AI-powered support chatbot capable of retrieving and understanding company documentation using Qdrant for vector search. The solution provides real-time responses based on the latest documentation updates, enhancing both customer and internal support efficiency.
Key achievements include:
Technologies used: Groq • OpenAI Whisper • ElevenLabs • Gemini • GCP Cloud Run • Next.js • Firebase • TypeScript • LangFlow • Qdrant • FastAPI
I developed a sophisticated system for automating responses to reviews on Google Places. The solution analyzes each review's language and contextual content to generate appropriate, personalized responses, streamlining business communication with customers.
Key achievements include:
Technologies used: FastAPI • Next.js • PostgreSQL • Qdrant Vector Database • LangChain • Pydantic AI • ChatGPT • GCP API & Services • OAuth2.0 • DeepSeek v3
manifōld AI
Copenhagen, Capital Region of Denmark · Remote
Own product vision and architectural design of the platform: define technology stack, drive key decisions on system design, security, and compliance.
Lead the creation and implementation of new services and microservice architecture, ensuring scalability, fault-tolerance, and enterprise-grade reliability.
Lead the evolution of the AI agents platform processing 2000+ events daily for enterprise clients with low-latency and high availability.
Deliver secure and compliant solutions, including high-compliance projects, third-party integrations, and data processing optimizations.
Partner with clients to transform business requirements into scalable technical solutions, bridging product and engineering.
Developed a new AI agent service for automated message processing that reduced agent development time from weeks to days, significantly accelerating time-to-client and improving delivery efficiency.
Technologies: Node.js • BullMQ • Pydantic AI • LangGraph • MCP • A2A • Python • Auth0 • FastAPI • React.js • Next.js • Azure • GCP • Redis • PostgreSQL • Qdrant
manifōld AI
Denmark · Hybrid
Designed and implemented a developer onboarding process, improving integration of new hires and accelerating team productivity.
Established CI/CD pipelines with automated security checks, enhancing code quality and enabling early detection of vulnerabilities.
Modernized legacy services step by step, delivering scalable architectures and improved customer UI/UX.
Optimized deployments, achieving annual infrastructure cost savings of €8,000+ through process redesign and resource-efficient tooling.
Built and maintained production-ready AI integrations (MCP, A2A), enabling reliable interactions with AI agents and boosting investor confidence.
Technologies: Node.js • BullMQ • Pydantic AI • LangGraph • MCP • A2A • Python • Auth0 • FastAPI • React.js • Next.js • Azure • GCP • Redis • PostgreSQL • Qdrant
Test-Driven Development with Python
Book, 2024
Designing Data-Intensive Applications
Book, 2024
AWS Certified Cloud Practitioner
AWS 2024
Open to discussing technical leadership opportunities, architectural challenges, or innovative AI-powered solutions. Let's explore how we can build something impactful together.