From fceab3c97c3f93cca09b8e00a3587f3839c3a16d Mon Sep 17 00:00:00 2001 From: artale Date: Tue, 7 Jul 2026 13:33:48 +0200 Subject: [PATCH] docs(tac): add NVIDIA ComputeX agent ecosystem appendix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Folds the Adel + Chris Murphy talk (youtube.com/watch?v=SVWmuJx0hHM) into loop_engineering.md. Covers the third inflection point (ChatGPT → DeepSeek → OpenClaw autonomous agents), agent = model + harness, OpenShell secure runtime with policy-based egress and secrets management, AIQ multi-agent deep research blueprint (intent router → orchestrator → specialists → synthesizer at 50% lower cost), NemoClaw blueprints, Nemo Relay observability with ATIF traces, and the Verified Skills pattern. Maps each component to TAC's existing factory infrastructure. --- plans/meta-prompts/loop_engineering.md | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/plans/meta-prompts/loop_engineering.md b/plans/meta-prompts/loop_engineering.md index 83bb193..5dd70f5 100644 --- a/plans/meta-prompts/loop_engineering.md +++ b/plans/meta-prompts/loop_engineering.md @@ -256,3 +256,25 @@ Reference: model-routing cost-savings tutorial (YouTube `SUZwYV5JYBM`), Coinbase - The brainstorms/spec skills naturally produce the architect's spec → handoff to builder pattern - Third-party harnesses (Cursor auto mode, Not Diamond) handle this automatically if configured - The Agent multiplexer note's orchestrator can route planning to one pane (Tier 1) and execution to another (Tier 2) + +## NVIDIA agent ecosystem — Neotron, OpenShell, Hermes + +Reference: NVIDIA ComputeX 2026 keynote — Adel (Senior Director AI Agents) and Chris Murphy (PM NemoClaw/OpenShell) at YouTube `SVWmuJx0hHM`. + +**Third inflection point:** ChatGPT (text gen, 2022) → DeepSeek (reasoning, 2024) → OpenClaw (autonomous self-evolving agents, Jan 2026). We're in the autonomous agent era. + +**Agent = model + harness.** The harness (tooling, orchestration, APIs, data sources) is as important as the model. NVIDIA's contribution spans the full stack. + +**Key architectures for TAC:** + +| Component | What it does | TAC relevance | +|-----------|-------------|---------------| +| **OpenShell** | Open-source secure agent runtime — policy-based egress control, secrets management outside sandbox, human-in-the-loop policy review | Natural complement to TAC's insider-threat model. Agent never sees real keys. Already being built into Windows and Ubuntu OS. | +| **AIQ Deep Research** | Multi-agent system: Neotron Nano (intent router) → GPT 5.2/Opus (orchestrator+planner) → Neotron 3 Super (5-6 specialist researchers) → synthesizer. #1 on Deep Research Bench at 50% lower cost. | Reference architecture for TAC's model routing. Intent router (small/cheap) decides workflow, frontier handles planning, mids handle execution. | +| **NemoClaw Blueprints** | Open-source reference implementations combining agent + OpenShell + models/tools. Hermes and OpenClaw variants. Customizable starting points, not end products. | TAC's factory already runs Hermes (3 variants) and OpenClaw. OpenShell sandboxing could be added around existing agents. | +| **Nemo Relay** | Agent observability + optimization. ATIF format traces, learns call patterns, hints Anthropic caching strategy, feeds KV-cache hints to Dynamo. ~30% token cost reduction. | TAC's model routing could integrate learned call-pattern hints. The ATIF trace format is human-readable and cross-platform. | +| **Verified Skills** | NVIDIA CUDA-X libraries packaged as plain-English skill manuals, scanned for vulnerabilities, evaluated across harnesses/models, cryptographically signed. | Patterns TAC can apply: skill metadata (what a tool is good/bad for), signature verification at runtime. | + +**ServiceNow case study:** 90% of L1 tickets resolved by autonomous agents using AIQ blueprint. System of models (frontier for orchestration, open for specialist researchers) with human-in-the-loop for escalation. + +**Practical takeaway:** TAC already aligns with the system-of-models pattern (see Model routing appendix). The next step is OpenShell-style policy gating around the Hermes/OpenClaw agents in the factory — sandbox the agents that have access to internal data sources (email, forgejo, git-proxy deploy) and keep human approval on policy changes.