AI & Software Architecture Consultant
Remote, EU timezone
- Architect and deploy privacy-first local AI products, enabling secure document intelligence where sensitive data is processed entirely on-device or on-premise, with no third-party data exposure by design
- Design multi-agent workflows and backend architectures using Python (Django, FastAPI) and containerized infrastructure across cloud providers, defining scoped agent permissions and output-validation steps for reliable autonomous operation
- Advise clients on infrastructure strategy, evaluating managed services (AWS RDS, GCP Cloud Run) against self-managed container deployments (DigitalOcean) across cost, operational overhead, and data-residency requirements
- Lead end-to-end integration of open-weight LLMs (via OpenAI SDK, Ollama, and MLX for Apple Silicon inference) into production environments, selecting and optimizing models for available RAM and latency budgets
- Apply threat modeling and zero-trust principles across delivered systems, with secure secrets management in CI/CD pipelines and scalable data layers in PostgreSQL and MongoDB