Lead UI Designer — Seldon
Seldon — Lead UI / Product Designer
Seldon is a leading MLOps company enabling enterprises to deploy, monitor, and govern machine learning models at scale across complex production environments.
At Seldon, I led the UI design of the company’s core SaaS platform, working at the intersection of machine learning infrastructure, DevOps practices, and enterprise software design. The platform served data scientists, ML engineers, and platform teams operating in Kubernetes-based environments — users with deep technical expertise but very limited time and high demands for clarity and reliability.
My role extended far beyond interface styling. I was responsible for shaping how complex ML workflows were understood and navigated inside the product. This meant translating abstract concepts such as model deployment strategies, inference graphs, monitoring pipelines, drift detection, and explainability tooling into structured, intuitive, and production-ready user experiences.
Key Contributions
End-to-end product design ownership
I worked closely with product managers, engineers, and leadership from early discovery through to shipped features. This included:
- Defining information architecture for model lifecycle management
- Designing workflows for deployment, monitoring, alerting, and governance
- Creating system-level UI patterns to support highly technical configuration tasks
- Supporting engineering during implementation to ensure fidelity and usability
Translating ML complexity into usable interfaces
Machine learning infrastructure is inherently complex. I focused on reducing cognitive overload by:
- Designing progressive disclosure patterns for advanced configuration
- Structuring multi-step deployment flows with clear validation states
- Improving visibility of system status, logs, and model health
- Creating clearer visual hierarchies for inference graphs and pipeline components
The goal was not simplification for its own sake, but clarity without loss of capability.
Platform-wide design system evolution
I redesigned and matured Seldon’s core design architecture, introducing:
- Consistent UI components and interaction patterns
- Structured layout principles for dense data environments
- Documentation and reusable design assets in Figma
- Alignment between brand expression and product experience
This created stronger cohesion across the platform and reduced ambiguity in future feature development.
Research-led usability improvements
Through collaboration with users and internal stakeholders, I identified friction points in core workflows — particularly around model configuration, monitoring interpretation, and debugging. These insights informed iterative refinements that made the platform easier to reason about and reduced avoidable configuration errors.
Mentorship and design quality standards
As part of the design leadership, I mentored other designers, introduced critique rituals, and raised quality benchmarks across interaction, accessibility, and consistency. I helped embed design more deeply into product decision-making rather than treating it as a downstream visual layer.
Outcome
My work at Seldon helped define a more coherent and approachable experience within a highly technical MLOps ecosystem. By aligning product strategy, engineering constraints, and user needs, we evolved the platform into a system that felt structured, dependable, and scalable — enabling enterprise teams to deploy and manage machine learning models with greater confidence.
