Explore my portfolio — services, professional background and live project walkthroughs.
I work in the Founder's Office at Crossing Hurdles, connecting AI project requirements with the people, workflows and tools needed to deliver them.
My experience spans LLM response evaluation, physical AI video-data programs, multilingual contributor operations, and building application platforms and recruiter dashboards. I bring a Mathematics background and a hands-on approach to moving projects from an initial brief to a usable workflow.
These are original portfolio demonstrations, separate from employer or client systems.
| Project | What it demonstrates |
|---|---|
| LLM Evaluation Lab | Reproducible aggregation of human-scored responses, weighted rubrics, critical-failure checks and transparent reports. |
| AI Contributor Operations Demo · Dashboard · Planning lab | Fictional brief-to-delivery planning, workload calculations and reviewed Markdown handoffs, alongside a submission-and-review dashboard. Rules-based, not live AI inference. |
| Responsive React Landing Page · Live demo | A responsive interface with accessible interactions, a local-only demo brief and documented implementation. |
Each repository includes a concise case study with design decisions, a walkthrough and limitations. The browser demos also include desktop and mobile screenshots.
- Translate an unclear brief into a defined scope and useful deliverables.
- Keep observations, assumptions and recommendations separate.
- Design workflows that make ownership and the next action clear.
- Test the work, document its limits and protect private information.
The projects use fictional or synthetic material, not private datasets or company source code. AI-assisted development is disclosed in the individual repositories, alongside setup instructions and validation notes. Synthetic examples are not presented as model benchmarks or client results.
Interested in thoughtful collaboration around AI evaluation, data workflows and practical software tools.