Praxis CI
c396ded395
TASK-07-01: server/cohort/aggregator.py — aggregate_session with k-anon write-time suppression (D-034, K_ANON_THRESHOLD=10), idempotent upsert, 7-day rolling window, multiple metrics (sessions_count, active_learners, gate_open_rate, median_mastery_score, rubric_criterion_means, failure_mode_frequency, branch distribution). No PII in aggregates (D-031). TASK-07-02: server/cohort/hook.py — on_session_end fire-and-forget (D-054), no-op when no Postgres, failures log + nightly reconciles. TASK-07-03: server/cohort/nightly.py — NightlyScheduler in-process asyncio loop, 03:00 CT (America/Winnipeg approx), reconcile from mastery_gate_events, R-DASH-04 failure handling. TASK-07-04: session_recorder.py — chain aggregation hook after mastery flow via asyncio.create_task (parallel, off voice path, D-054). TASK-07-05: tests/test_cohort_aggregation.py — k-anon threshold (9/10/11), idempotent, 7-day window, metrics, no PII. TASK-07-06: tests/test_cohort_nightly.py — scheduler timing, reconciliation, hook-failure+nightly recovery, R-DASH-04. G-038 (binding): differencing-attack test — 10 learners window A, 9 in B, verify dropped learner cannot be isolated (B suppressed, value=NULL). ---ci--- project: praxis phase: 2 milestone: v0.4 status: execute persona: backend-engineer task: 07-01..07-06 requirements: covered: [REQ-MT-02, REQ-NFR-DASH-02, REQ-NFR-DASH-01] ---/ci---
Praxis — v0.1 Foundation
Voice-first AI apprenticeship platform. v0.1 is a tech-validation harness (per G-008) for the minimal viable voice loop: a single learner speaks to an AI tutor playing a Customer Service role-play scenario, hears a <600ms-latency response, receives an end-of-session coaching debrief, and has the session logged to SQLite.
Status
Phase 1 (minimal viable voice loop) — code-complete, pending live API keys for runtime verification.
Stack
- Orchestration: Pipecat (D-017) with Silero VAD + interruptibility
- ASR: Deepgram Nova-3 streaming (D-013)
- LLM: Ollama Cloud direct API (D-020) —
gemma4:cloud(role-play) +deepseek-v4-flash:cloudno-think (debrief) - TTS: Cartesia Sonic (primary, D-014) / Piper (self-hosted, R4 mitigation) — behind an interface
- Client: React + Vite + WebRTC (Pipecat client SDK, D-015)
- State: SQLite
praxis.db(D-007, single hardcoded learner, no auth)
Layout
server/ Pipecat pipeline, services (TTS/LLM/Guardrail interfaces), scenario runtime, adapters
client/ React + Vite + WebRTC learner surface
scenarios/ YAML scenario definitions (D-018)
db/ SQLite schema, migrations, async store
scripts/ Latency probes (R1-R4), e2e smoke
tests/ Unit + e2e
docs/ Latency report, debrief templates
Quickstart
- Copy
.env.example→.env, fill inDEEPGRAM_API_KEY,CARTESIA_API_KEY,OLLAMA_API_KEY. - Install server deps:
pip install -e ".[dev]" - Install client deps:
cd client && npm install - Run probes:
python scripts/probe_deepgram.py(etc.) - Run server:
python -m server - Run client:
cd client && npm run dev
See docs/latency-report.md for the R1-R4 spike status and TTS decision.