Case study · AI/ML platform
Orchestra: designing an MLOps platform for GPU, Kubernetes and LLM workloads
An MLOps platform for starting and governing AI workloads: GPU allocation, Kubernetes orchestration, live monitoring and LLM inference.
- Role
- Sole designer
- Focus
- MLOps, monitoring
- Tools
- Figma
- Sector
- AI/ML platform
- Year
- To be added

The brief
Give teams one place to start AI workloads, allocate GPUs, orchestrate them on Kubernetes, monitor them live and serve LLM inference.
My role
I was the sole designer on this project. The screen here is the projects dashboard.
Details coming soon.
Scope
- Focus: MLOps, monitoring
- Tools: Figma
Details coming soon.
Process
Details coming soon.
Key decisions
Details coming soon.
Outcome
Details coming soon.
There is no public write-up of this project yet. Ask me about it.
Working on something similar?
Send a short note about the product, the team and the timing. Email is the quickest way to reach me.