Sauti Mtaani Nairobi
Bilingual civic feedback platform with human-reviewed AI
- Status
- Live web app
- Engagement
- Civic platform for the Office of the Nairobi Woman Representative
- Role
- Full-stack & AI engineer
The problem
Residents need a clear way to submit structured feedback, and a constituency office needs aggregated insight without automating decisions that require human judgement.
What was built
A bilingual resident-feedback platform that aggregates submissions and uses AI to suggest themes, with every AI suggestion reviewed by staff. PII is scrubbed before any AI call, and there are no automated eligibility decisions.
Architecture and engineering decisions
A Next.js and TypeScript application on Supabase, with consent and access controls, a PII-scrubbing step before Groq model calls, and a staff review queue for AI-generated suggestions.
Audience
Nairobi residents and constituency office staff reviewing feedback themes.
Technical highlights
- Human review of every AI suggestion
- PII scrubbed before model calls
- Bilingual resident feedback
- Aggregation and theme support for staff
- No automated eligibility decisions
My responsibilities
- Full-stack implementation
- AI workflow design
- Privacy safeguards
- Staff review flow
Why it matters for a client
Shows privacy-aware AI in a civic workflow where human judgement and resident trust are central.
Constraints and trade-offs
Advisory AI only; staff approve suggestions. The platform is not an eligibility or benefits determination system, and no political persuasion framing is used.
Privacy and accessibility
Consent flows, PII scrubbing before AI calls, and staff approval of AI suggestions. No claim of formal security certification.
Verified evidence
Live at sautimtaani.site and described on the owner portfolio as using human-reviewed AI and PII scrubbing. Not independently security-audited for this site.
Limitations
No public usage, adoption or decision-quality metrics. Civic context only; no endorsement or official determination implied.