Live demo
NPC Coach. Every new patient call graded against the playbook.
Every new-patient phone call gets transcribed and graded against the practice's documented playbook, producing a coaching report that quotes the exact moment a coordinator missed a non-negotiable, next to their own 90-day average.
Walk through it, six steps
Two ingestion paths, one call record
Outbound calls arrive as MP3 drops on SFTP; inbound calls arrive from the call platform with a transcript already attached, which the pipeline reuses instead of paying to transcribe twice. Both normalize into one record, claim an idempotency key, and create a visible run row before any filtering, so no call can silently disappear.
The problem.
New-patient coordinators are the practice's first advocates, and their calls set the tone for everything after. But nobody could listen to a thousand calls a month, so coaching was anecdotal. The design constraint came from the top: correctness beats conversion, and booking rate is explicitly not the success metric.
- 285 → 40 recordings ingested, relevant calls graded
- 1k to 5k calls/mo 5 to 10 minutes each, EN + ES
- 1,000+ across ~19k lines of Python
- 16 days including the React dashboard
How it fits together.
Ingest
- SFTP MP3 drops + call-platform webhooks
- One normalized call record, idempotency claims
- Run rows created before any gate: nothing disappears
Understand
- Speech-to-Text v2 with diarization (outbound)
- Platform transcripts reused (inbound), cost deduped
- Call classification: new patient vs everything else
Score
- Gemini structured output with response schema
- 6 criteria, 3 hard gates, weighted buckets 40/40/20
- Verbatim evidence quotes with a relevance gate
Deliver
- React dashboard: runs, steppers, logs, analytics
- Email reports to coordinator + manager
- Firestore state · BigQuery audit · Terraform · ports-and-adapters demo mode
Where it landed.
A thousand-plus monthly calls now get the same rubric a human coach would apply on their best day, with receipts. The ports-and-adapters design means the whole pipeline runs on synthetic data in demo mode, which is exactly what this tour is. This is a PHI workload in production: no real transcripts, recordings, or caller details appear here, and coordinators are anonymized as C-1 through C-8.