# TARA

- **Event:** [Built with Opus 4.6: a Claude Code hackathon](https://cerebralvalley.ai/e/claude-code-hackathon)
- **When:** Feb 10 at 12:00 PM – Feb 17 at 10:00 AM (EST)
- **Where:** Location TBA
- **Placement:** Finalist
- **Team:** [Kyeyune Kazibwe](https://cerebralvalley.ai/u/KyeyuneKazibwe)
- **GitHub:** https://github.com/Kye256/tara-transport-assessment
- **Demo video:** https://www.youtube.com/watch?v=GFCrXehS1DE
- **Gallery:** https://cerebralvalley.ai/e/claude-code-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/claude-code-hackathon/hackathon/gallery/56

Africa needs billions annually in transport infrastructure but most projects fail at feasibility. Road appraisal takes weeks of specialist work and most agencies lack capacity. In Uganda we spend over $1B yearly on roads with $20M on planning and the process is manual and inefficient.
TARA transforms this by allowing us to use dashcam footage. Claude Opus 4.6 analyses every frame to give surface condition and roadside activity. It segments the road, selects interventions with Uganda-calibrated costs, and runs full cost-benefit analysis with NPV, EIRR, BCR, sensitivity and deterioration modelling.
What makes TARA different: it asks who benefits. The equity module identifies pedestrians without footpaths, school children on the carriageway. It shifts appraisal from economics to human-centred analysis.
Built with Claude Opus 4.6 for vision, condition narratives, sensitivity and equity. Tested on real dashcam footage near Kampala. Open source.

---

Markdown version of https://cerebralvalley.ai/e/claude-code-hackathon/hackathon/gallery/56. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt