groundy

Groundy — independent coverage of developer tools, infrastructure, and platforms

infra

daVinci-kernel shifts the RL kernel bottleneck from reward shaping to skill libraries

The daVinci-kernel preprint argues RL-tuned CUDA kernels plateau because skill libraries are mis-architected, not because rewards are mis-shaped. It co-evolves skill.

6min·08:05·jul 30
agents

Cloudflare Precursor: Behavioral Detection for AI Agents

Cloudflare Precursor reportedly shifts AI agent detection from spoofable headers to continuous behavioral signals. This article verifies the claims and outlines operator.

6min·15:05·jul 30
devtools

Provenance as a CI Gate: Attributing Agent-Authorship in Code

Git credits the merger, not the model. F(AI)2R proposes a PROV-O provenance graph gated by CI to record agent authorship and human verification, preventing blame collapse in.

devtools

OHTTP CLI: Stateless Privacy for Agents vs VPN and Tor

Cloudflare's OHTTP protocol splits request identity from content across two parties. This CLI approach offers stateless privacy for agents, shifting key management burden to.

models

Kimi K3 on M1 Max: Bandwidth, Not Capacity, Limits Local MoE Inference

Running Kimi K3 on an M1 Max proves local MoE feasibility via expert offloading, but unified memory bandwidth caps throughput. Treat this as a prototyping tool, not a serving.

policy

Why Written AI Policies Fail to Control Agent Behavior

HANDBOOK.md benchmark shows frontier agents follow 124-page policies on only 36.2% of trials. Governance requires runtime enforcement, not documentation.

agents

Why Multi-Agent LLM Delegation Concentrates Risk

Splitting tasks across planner and worker agents concentrates risk because principals cannot observe hidden actions. Base models defect. Teams must add verifiable execution.

models

Kimi Linear Cuts KV Cache 75% but Recall Remains the Binding Constraint

Kimi Linear cuts KV cache 75% and boosts decode 6x for 1M-token contexts, but recall stays the binding constraint. Route memory-bound jobs only after benchmarking in-context.

featurededitor's picks
most popularlast 6 days · by traffic
tophighest rated
feedall 650 articles · reverse chronological
  1. jul 30infradaVinci-kernel shifts the RL kernel bottleneck from reward shaping to skill libraries
  2. jul 30agentsCloudflare Precursor: Behavioral Detection for AI Agents
  3. jul 30devtoolsProvenance as a CI Gate: Attributing Agent-Authorship in Code
  4. jul 30devtoolsOHTTP CLI: Stateless Privacy for Agents vs VPN and Tor
  5. jul 30modelsKimi K3 on M1 Max: Bandwidth, Not Capacity, Limits Local MoE Inference
  6. jul 30policyWhy Written AI Policies Fail to Control Agent Behavior
  7. jul 29agentsWhy Multi-Agent LLM Delegation Concentrates Risk
  8. jul 29modelsKimi Linear Cuts KV Cache 75% but Recall Remains the Binding Constraint
  9. jul 29policyWhy Vendor Model Cards Fail Clinical Ethics Procurement
  10. jul 29agentsHarness vs Scaffold: Why Claude Code and LangGraph Are Not Interchangeable
  11. jul 28devtoolsFine-Tuning vs RAG for Internal APIs: StarCoder2 Constraints
  12. jul 28agentsMCP Tool Discovery Moves From Hardcoded Config to Runtime Agent Search
  13. jul 28infraCalibrated LLM Monitoring: Conformal Prediction with Drift Detection
  14. jul 28devtoolsMellum2 Unverified: Why MoE Active Parameters Matter More Than Total Size
  15. jul 28agentsCode-as-Action Agents Beat GAIA But Require Runtime Sandboxing
  16. jul 28policyTRIDENT Benchmark: LLM Safety Gaps in Finance, Medicine, and Law
  17. jul 28modelsWhy Chat Leaderboards Do Not Predict Image Quality
  18. jul 27devtoolsPyPI Wheel Reproducibility: 15% Byte-Identical, 79% Source-Equivalent
  19. jul 27modelsKimi K3 Procurement: Governance Review Over Phantom Government Assessments
  20. jul 27policyEU AI Act Traceability: Why ML Pipelines Fail Conformity Assessment
  21. jul 27infraCloudflare AI Crawler Controls: Block, Charge, or Allow Bots Per Route
  22. jul 27agentsWhy Agent Security Tests Must Audit Full Trajectories, Not Single Turns
  23. jul 26agentsx402 Per-Call Payments: Agent Wallet Custody and Replay Risks
  24. jul 26modelsDeepSeek Compute Leak: Why Open-Weight Routing Needs a Swap Path
  25. jul 26modelsContext Ordering Beats Window Size for Long-Context Agents
  26. jul 26devtoolsCHRONO-RESOLUTION: npm, PyPI, and crates.io lockfile drift measured at release points
  27. jul 25infraPostgres LISTEN/NOTIFY Scales: When to Drop Redis for Job Fan-Out
  28. jul 25devtoolsCLI-Tool-Bench: Why Patch Leaderboards Fail for 0-to-1 Code Generation
  29. jul 25policyImplicit Bias in LLMs Passes NYC and EU Audits
  30. jul 25agentsCodeRabbit Review Study: 56% Rejection Rate Demands Targeted Scoping
  31. jul 24modelsDiffusion LLMs: Training Cost, Not Parallel Decoding, Drives Deployment
  32. jul 24infraTailscale on Azure: Measure Direct vs DERP Routing to Control Latency and Egress
  33. jul 24infraAccelerate vs Megatron Core: The Model Size Curve for Distributed Training
  34. jul 24agentsLLM Agents Ignore Mid-Flight Halt Signals: 0 of 40 Trials Stopped
  35. jul 24modelsOpen-Weight Routers vs Fable 5: The Routing Math That Actually Matters
  36. jul 24agentsAgent-First CLIs: Why GitHub, npm, and PyPI Must Publish Machine-Readable Contracts
  37. jul 23policyEU Driver Monitoring: GDPR Compliance Without Consent
  38. jul 23infraWhy cgroups, not permission prompts, bound AI agent CPU and memory
  39. jul 23modelsDeepSeek-V4 1M Context vs RAG: Why Retrieval Stays
  40. jul 23modelsQwen-Image-3.0 Does Not Exist: Why Self-Hosting Image Models Is Premature
  41. jul 22policyEU AI Act bans emotion AI in schools, but permits it where models fail
  42. jul 22agentsMCP and AGENTS.md Standardize Context, Not Agent Coordination
  43. jul 22industryAI Search's One-Answer Rule: When Better Content Makes Search Worse
  44. jul 22agentsCursor's Swarm Math: When Cheap Agents Save Money and When They Fail
  45. jul 21agentsRuntime monitoring beats alignment for agent-to-agent coercion
  46. jul 21infravLLM Configs Shift Energy, Latency, and Accuracy: A 9,000-Run Study
  47. jul 21policyHuggingFace vs GitHub Models vs Replicate: Policy Compliance for Uploaders
  48. jul 20modelsKimi K3: 2.8T Parameters, MoE Routing, and Self-Hosting Reality
  49. jul 20modelsKimi K3 vs Qwen3.8 Max: Routing Strategy for July 2026
  50. jul 20agentsCloudflare's Agent Stack: Edge Trust, Identity, and Metering
load older →