groundy

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

infra

GPU Memory Explained: Why LLM Throughput Collapses When VRAM Runs Out

LLM decode speed tracks memory bandwidth, not TFLOPS. When VRAM fills, reads shift to PCIe, causing a throughput cliff. Use this bandwidth budget worksheet to plan KV cache.

6min·10:50·aug 23
policy

DiverValue-Bench: Measuring LLM Value Divergence Across 74 Markets

DiverValue-Bench scores LLM value divergence across 74 markets using 23,763 instances. It reveals that passing bias audits does not ensure value alignment, forcing deployers.

6min·14:33·aug 23
models

Prompt Injection in 3D Scenes: The Attack Surface Multimodal Agents Ignore

arXiv 2602.07104 shows 3D object placement injects instructions into multimodal agents, bypassing text and image filters. Teams must gate modalities and scope actions to.

infra

Task-Based OAuth Consent: Scoping AI Agent Permissions Per Action

Cloudflare's task-based OAuth consent moves authorization from install time to runtime. Compare blast radius, revocation, and fatigue to redesign agent API security.

industry

Audio Token Compression: Cutting Voice LLM Inference Costs

A preprint claims 3x audio token reduction for voice agents, but quality drops without LoRA finetuning. This guide maps the cost, latency, and build-vs-buy tradeoffs for.

agents

Anthropic A/B Tests Claude Code Effort Levels: Your Agent Is the Control Group

An unverified X post claims Anthropic is A/B testing reduced effort in Claude Code. This guide covers eval canaries, variance monitoring, and contract tests to catch silent.

agents

Self-Hosted Coding Agents: The Safety Burden You Inherit

Self-hosting coding agents transfers safety ownership to your team. Benchmarks show automated detection fails on novel sources, making human review the true bottleneck.

models

Why LLM Log Anomaly Detection Pages You for Nothing

arXiv 2608.17965 shows LLM log detectors are overconfident in wrong verdicts. Gate paging on calibrated confidence, not raw F1, to prevent alert fatigue and ignored critical.

featurededitor's picks
most popularlast 6 days · by traffic
tophighest rated
feedall 691 articles · reverse chronological
  1. aug 23infraGPU Memory Explained: Why LLM Throughput Collapses When VRAM Runs Out
  2. aug 23policyDiverValue-Bench: Measuring LLM Value Divergence Across 74 Markets
  3. aug 23modelsPrompt Injection in 3D Scenes: The Attack Surface Multimodal Agents Ignore
  4. aug 23infraTask-Based OAuth Consent: Scoping AI Agent Permissions Per Action
  5. aug 23industryAudio Token Compression: Cutting Voice LLM Inference Costs
  6. aug 23agentsAnthropic A/B Tests Claude Code Effort Levels: Your Agent Is the Control Group
  7. aug 22agentsSelf-Hosted Coding Agents: The Safety Burden You Inherit
  8. aug 22modelsWhy LLM Log Anomaly Detection Pages You for Nothing
  9. aug 22modelsDeepSeek v4 Flash Vision: Routing Images Without Verified Pricing
  10. aug 22agentsClaude Code Weekly Limits Promo Ends in August 2026: Budgeting for Agent Teams
  11. aug 21modelsDeepSeek 32B on RTX 3090: Tokens per Second by Quant and Context
  12. aug 21modelsPTXBench: LLMs Can Port GPU Kernels, But Not Beat Tuned Libraries
  13. aug 21agentsVibe Coding vs Control: How Developers Actually Used AI Coding Agents
  14. aug 21policyAndroid Privacy Policies vs. Runtime Logs: A 0.4% Alignment Study
  15. aug 21industryLLM Data Center Control: Why Advisory Beats Closed-Loop
  16. aug 21infraV8 Isolates vs MicroVMs vs Wasm: Where Spectre Still Draws the Line
  17. aug 20modelsCan LLMs Reuse Another Model's KV Cache? What Cross-Model Transfer Shows
  18. aug 20infraFine-Tuning DeepSeek Without NVIDIA: What the Ascend SuperPOD Run Shows
  19. aug 20agentsMulti-Agent or Single-Agent LLM: What Skill Distillation Actually Costs
  20. aug 20devtoolsTraining a Personal Coding Assistant: GPU Cost vs a Copilot Seat
  21. aug 20agentsMulti-Agent LLM Systems Drift Into Misaligned Communication Over Long Horizons
  22. aug 19infraCloudflare WebMCP: The Security Baseline for Agent-Ready Sites
  23. aug 19policyWhy Machine Unlearning Can't Certify GDPR Erasure
  24. aug 19agentsSizing Agent Memory: A Capacity Planning Rubric for Long-Horizon LLMs
  25. aug 19infraCloudflare AI Search vs Self-Hosted RAG: Where the Build-vs-Buy Line Lands
  26. aug 18devtoolsVCoT-Bench: Why AI Rust Verification Fails Merge Gates
  27. aug 18infraCloudflare H1 2026 DDoS Report: DNS Floods and Sizing Past 1 Tbps
  28. aug 18industryLLM Conflict-of-Interest Benchmark: Sponsor Bias as a Measurable Failure Mode
  29. aug 18policyRA-Bench: Why Deepfake Detectors Fail on Re-Shared Crisis Video
  30. aug 18agentsWhen Spec-First Agents Dismantle Invariants: A Governance Case Study
  31. aug 18devtoolsRust GPU Offload: arXiv 2608.13759 Analysis for Rust Teams
  32. aug 18agentsCloudflare Kitesurf: V8 Isolates vs Containers for Agent Browsing
  33. aug 18policyDario Amodei on AI Regulation: Frontier Labs as Their Own Lobbyists
  34. aug 18agentsClaude Code Skills vs Model Weights: Where Should Agent Skills Live?
  35. aug 17devtoolsCursor Origin vs GitHub: The Real Cost of Switching Repo Hosts
  36. aug 17devtoolsTreat AI Autofix as Untrusted Input: Merge Gates and CI Scoping
  37. aug 01modelsDeCRIM: Decompose Constraints to Stop Silent Drops in Agent Outputs
  38. aug 01devtoolsKimi K3 Local Inference: Why 2.8T Parameters Break the Consumer RAM Floor
  39. jul 31modelsOperator-Level Triage for Silent Mixed-Precision Instability
  40. jul 31modelsBeyondUncertainty: Weak Confidence Signal for RAG Routing, Not Calibration
  41. jul 31policyPublic Sector AI Procurement Must Shift from Model Certification to Task Authorization
  42. jul 30infradaVinci-kernel shifts the RL kernel bottleneck from reward shaping to skill libraries
  43. jul 30agentsCloudflare Precursor: Behavioral Detection for AI Agents
  44. jul 30devtoolsProvenance as a CI Gate: Attributing Agent-Authorship in Code
  45. jul 30devtoolsOHTTP CLI: Stateless Privacy for Agents vs VPN and Tor
  46. jul 30modelsKimi K3 on M1 Max: Bandwidth, Not Capacity, Limits Local MoE Inference
  47. jul 30policyWhy Written AI Policies Fail to Control Agent Behavior
  48. jul 29agentsWhy Multi-Agent LLM Delegation Concentrates Risk
  49. jul 29modelsKimi Linear Cuts KV Cache 75% but Recall Remains the Binding Constraint
  50. jul 29policyWhy Vendor Model Cards Fail Clinical Ethics Procurement
load older →