Tracking emerging technology, dissecting system architecture, and publishing what I actually build and evaluate.
Start with implementation work, then move to research updates or concept guides.
Agents, RAG, harnesses, and platforms: implementations, evaluation, and trade-offs.
New models, frameworks, and industry shifts, with fast context and technical judgment.
Browse researchCurated paths for building a coherent model of agents, RAG, and harness engineering.
Browse concept guidesAnthropic estimates that robots can perform 74% of US physical tasks in some settings, yet are cost-competitive for only 0.3% of work today. This analysis explains the tiers, cost assumptions, and implications for deployment decisions.
A closer look at H Company’s cross-interface agent and long-horizon harness, its benchmark claims, public traces, and the licensing split between checkpoints.
NVIDIA Kumo Tabular reframes prediction as in-context learning: a model pretrained on synthetic tables predicts new rows from labeled examples. We examine its method, vendor-reported leaderboards, and enterprise validation requirements.
A technical analysis of OpenShell's inspectable agent-runtime controls and NVIDIA's BlueField-4/Sentry reference design, separating their responsibilities from the efficacy evidence still needed.
An engineering analysis of Target's lexical and vector retrieval, attribute controls, weighted interleaving, and the evidence limits behind its reported gains.
How Microsoft’s run-assert-eval connects Clarity risk discovery, ASSERT behavior tests, and ACS runtime policy, while measuring unsafe behavior separately from over-refusal.
A close look at how Docker Sandbox Kit v3 places agent network, credential, and mixin declarations in an OCI artifact—and why a descriptor remains a request whose enforcement depends on the runtime.
Anthropic's low-stakes employee book exchange found that preference estimates constrained outcomes more than bargaining rules, making preference understanding a separate test for delegated agents.
A close look at AlloyDB for agents’ read-only nodes, dedicated Colossus segments, and elastic compute pool, including isolation, freshness, cold starts, cost, and transaction boundaries.
A practical look at how Cloudflare Turnstile Spin installs, repairs, or migrates a widget with human approval, while Siteverify, credential isolation, and replay testing close the server-side security loop.
A layered design for running AI-generated code in a separate AWS account, with per-job credentials, S3 endpoint policies, DNS allow-listing, and continuous exfiltration tests.
XYEval shows how a plausible but misplaced user suggestion can lower agent success; the practical response is to verify the goal, then explain a better path with evidence.
MobileCybench replays Android agent exploits in an isolated environment, then checks trusted state with executable probes to see which security properties were violated.
Trace how AWS’s multi-account AgentCore Gateway pattern separates user JWT authorization through Cedar from downstream M2M OAuth, and identify the sample’s production hardening gaps.
A focused engineering reading of Ouroboros v7.4.4: how durable identity, memory, self-modification, and managed subagents can be connected to SHA256, SBOMs, GitHub build provenance, and smoke receipts before deployment.
A focused engineering reading of RSIAgent's curriculum, actor, verifier, and broad-to-deep exploration loop, with a careful audit of frozen experience, benchmark reporting, and reproducibility limits.
Anthropic proposes three measurements for AI-led R&D: automation, agent oversight, and safety compute; this article separates its internal self-report from methodology limits and cross-lab comparability.
A focused engineering reading of gh-aw v0.89.17: how log audits, MCP Gateway and firewall boundaries, grading, model-cost signals, and incident monitoring move Agentic CI beyond a demo.
Try a broader keyword or clear the current filters.
For speaking invitations, internal engineering sessions, or architecture exchange, see the topics and public work I can bring into the conversation.
Speaking & contact
