AI / ML Engineer
Pol Fernández
Building intelligent systems
for a smarter tomorrow.
- Building software
- +5
- Systems in production
- 20
- Working languages
- DE EN ES
About
Systems built to run unattended.
I design and build AI systems for production use: multi-agent orchestration, retrieval over controlled document sets, and the integration work that connects them to the systems a company already operates.
The larger part of the effort goes into what determines whether a system survives real use. Access control, error handling, observability and a complete audit trail are part of the specification, not of a later phase.
- Based in
- Zurich, Switzerland
- Focus
- AI orchestration and automation
- Stack
- Python, TypeScript, PostgreSQL, Linux
- Languages
- German, English, Spanish
Technologies
- Python
- TypeScript
- PostgreSQL
- Docker
- Kubernetes
- Linux
- Git
- GitHub Actions
- Azure
- Cloudflare
- Playwright
- n8n
- Claude
- LangChain
- Ollama
- Proxmox
Featured project
Pharma and laboratory robotics
Multi-agent orchestrator for test automation
A test suite for instrument software, originally written in a tool that is being withdrawn, covering equipment used in the handling of patient samples. The system coordinates seven specialised agents across fifty two tools and three model backends.
Every run is instrumented. Cost, duration and the changes produced are recorded, which allows each agent to be assessed against its measured contribution rather than against a demonstration.
- Specialised agents
- 7
- Tools
- 52+
- Model backends
- 3
- UC-01Order generator
- UC-02Log management
- UC-03Deck visualiser
- UC-04Protocol parser
- UC-05Pull request reviewer
- UC-06Knowledge consolidation
- UC-07Test orchestration
Selected work
Further projects, by sector.
-
Defence and optronics
Retrieval over classified documentation
Air-gapped deployment, source citation on every answer, and an audit trail of who queried what.
-
Health and wellness
CRM with access control enforced in the database
Row level security in PostgreSQL, with a policy test per role and field level encryption.
-
Pharma and laboratory robotics
Quality gates in the release pipeline
Containerised runners and a virtual machine snapshot per execution, so every failure is reproducible.
-
Research and prototyping
3D scanning and robotic arm control
Geometry acquisition, motion planning and a simulator for collision checking before execution.
-
Internal systems
Scheduled agents with cost reporting
Daily review of incoming signals, and automated transfer of time records between two employer systems.
Infrastructure
A production equivalent environment of my own.
Approximately twenty services run on dedicated hardware under my administration: single sign-on in front of all of them, a reverse proxy managing its own certificates, monitoring with alerting, and a six layer backup strategy verified by restore. Operational decisions are tested there before they are applied in a client environment.
- Services
- Approximately 20, each isolated in its own container
- Access
- Single sign-on, no service exposed by default
- Monitoring
- Alerting on failure, ahead of user impact
- Backups
- Six layers, verified by restore
Contact
Available for new projects.
Describe the problem you need solved. You receive a reply within one working day, including a direct answer if it falls outside what I can usefully contribute to.
Based in Au, Lake Zurich. Working in German, English and Spanish.