Restructure the 60s script to hit the judging tracks explicitly: real-time
multimodal perception (Gemini Vision + git fusion), live intervention voiced
by an on-device Gemma model (with Gemini TTS fallback + a pre-flight gate so
it's never debugged on stage), and the self-improvement stack narrated
(Atlas traces, vector recall, outcome-conditioned policy, 'Seen before'
escalation, recursive self-tuning). Adds a track-callout table and fallbacks.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LuV8W8oNYRsDWKoqK8Mkqc
Reframe the demo around the real shift: AI collapses the cost of writing
code (humans-with-AI now, agent swarms next), so the bottleneck moves from
engineering to coordination/management — and at agent scale no human can
track progress or avoid collisions by hand. Add a dedicated 1-minute live
script for showing real-time card+voice interventions with teammates, and
align the 4-min opening/close with the thesis.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LuV8W8oNYRsDWKoqK8Mkqc
Reframe README around the real bottleneck (human coordination, not
engineering ability) with the "five-minute meeting" cost story; rewrite the
demo script to match. Update gemini/livekit/mongodb/digitalocean specs to
reflect shipped code (Gemini Live agent, TTS, embeddings, Lyria via
Interactions API; current API routes; hermes_jobs collections). Add hermes.md.
Rename graph.md -> cont_learning.md. Remove outdated/dead docs (agent-learning
scaffolding, graph-discovery, handoffs, superpowers, idea/plan/demo-setup) and
the bundled LiveKit starter under examples/. Rewire CLAUDE.md doc-first gate
off the deleted PLAN.md.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LuV8W8oNYRsDWKoqK8Mkqc
Ramis