Open Source
Doory vs Penpot: Choosing a Self-Hosted Design Tool for Marketing Assets
Route template-driven marketing assets to a Doory pilot, keep design systems in Penpot, and use Canva for catalogs, noting Doory claims are project-reported.
The Groundy archive · Page 7 of 43
All articles
Browse Groundy's complete archive of 1027 articles on AI, developer tools and infrastructure. Page 7 of 43.
The archive
145–168 of 1027 articles · Newest first
Developer ToolsCodex CLI Now Runs a Self-Updating Daemon: How to Audit and Stop It
A single report suggests Codex CLI 0.157.0 spawns a self-updating daemon that bypasses package managers. Verify with pgrep and disable daemon_auto_start to maintain control.
- Ethics, Policy & Safety
Why AI Unlearning Audits Pass Models That Still Remember
Preprint evidence suggests output-level unlearning audits pass models retaining recoverable traces, requiring relearning probes and representation checks for valid erasure.
Infrastructure & RuntimeCan Idle Macs Become a GPU Cloud? What an Apple Silicon Cluster Delivers
A preprint shows idle Macs can pool for batch embeddings and Ollama completions, offering private inference that complements rented GPUs for latency-tolerant work.
Agents & FrameworksWhen Coding Agents Retry the Same Fix: What Repair History Changes
Coding agents retrying failed patches benefit from pre-loop context injection of repair history, a pattern supported by HAFixAgent and harness sensitivity studies.
Culture & SocietyWhen an AI Mathematician Tries to Write a Proof, Who Verifies It?
Three self-reported AI math repos lack expert verification, prompting a two-gate standard requiring full certificates or specialist attestation for citability.
Agents & FrameworksShould a Coding Agent Buy Its Own Domain? Where the Approval Belongs
Evidence suggests coding agents should gate spending via tool interfaces rather than prompts, as prompt-based budgets fail and vendor changelogs show structured refusals are a
Models & ResearchCan a LoRA Adapter Keep Learning? What Continual Fine-Tuning Costs
LoRA adapters can continue learning but risk overwriting prior knowledge. C-LoRA is vision-only, while LLM evidence remains preliminary. Use data mixing and rollback checks to
Developer ToolsWhen DuckDB 2.0 Speeds Up S3 Parquet Queries and When It Doesn't
DuckDB 2.0 alpha cuts S3 Parquet scan time from 18.8s to 7.7s, but gains depend on workload shape. Keep production on stable 1.5.6 and re-benchmark before migrating.
Culture & SocietyCockroach Labs Spent 5 Months Treating Bugs Like Patients: What Humans Keep
Cockroach Labs reports a five-month agent triage experiment. Independent studies show agents fail at reproduction, so adopt the review gates, not the self-reported volume.
Open SourceUpgrading to DuckDB 2.0: Which Workloads the Async I/O Speedup Helps Most
DuckDB 2.0 alpha async I/O speeds up large S3 reads by 2-3x per vendor benchmarks, but gains vanish for tiny files or local disk, requiring users to test their own workloads.
Agents & FrameworksRunning a Personal AI Agent on Cloudflare's Free Tier: What Fits
Talorys fits a personal AI agent on Cloudflare's free tier by using Durable Objects for state and Workers AI for chat, though quota limits and memory recall quality remain unv
SecurityLLM Jailbreak Scaling Laws: Sizing Red Team Budgets and Rate Limits
A preprint suggests jailbreak success shifts from polynomial to exponential growth with injection, implying red teams should test multiple effort levels and monitor search.
Open SourceHow to Point a Coding Agent at 400 Years of Archives: What to Verify
Antiquity uses a single GPU to generate archival leads, but findings remain unconfirmed until specialists verify scanned primary sources and known-event controls pass.
Infrastructure & RuntimePriority Queues or Budgets: How Ai2 Divides an Oversubscribed GPU Cluster
Ai2 reports that replacing priority queues with GPU budgets and fair-share allocation delivered 98% of owed hours, though evidence is self-reported and limited to one 30-day.
Developer ToolsHow to Evaluate AI Agents on Long Tasks: Public Benchmarks vs Your Own Environment
Public benchmarks offer coarse comparability, but long-horizon agent rollouts require custom environments with audited checkers and measured reset reliability.
Culture & SocietyLLM Taste Profiles vs Click History: Who Owns What You Like?
A single writer's LLM taste profile offers portability and data minimization, but privacy claims remain untested hypotheses lacking empirical validation against click-history.
Models & ResearchHow to Fine-Tune an LLM on Domain Data Without Losing General Skills
Preprint evidence shows domain fine-tuning causes 3.4-14.5% general skill loss. Mitigations like replay and routing reduce this, but require rigorous regression testing to be.
Developer ToolsPresidio vs Multilingual NER: Choosing a PII Redaction Approach
Presidio excels in English but fails in Russian, while trained models offer better multilingual coverage. Choose based on traffic language and build per-language evaluation.
Agents & FrameworksThe Cost of Agent Memory: What MemCalib's Benchmark Measures
MemCalib measures agent memory calibration, while other benchmarks quantify build costs and step savings, enabling per-workload keep, trim, or skip decisions.
Open SourceGitHub Forks Miss Upstream CVE Patches: How to Audit Yours
A preprint maps 2.2 million forks vulnerable to upstream CVEs, showing that standard alerting misses fork divergence. Auditing commit history is required to verify patching.
Ethics, Policy & SafetyAnthropic's 2026 Usage Policy: Who Can Stop Claude-Driven Physical Systems
Anthropic's 2026 policy requires Claude-connected hardware to have operator stop controls and disconnected fail-safes, creating a compliance deadline for physical systems.
Agents & FrameworksDocker Agent: Which MCP Tools and Container Permissions to Grant It
This guide maps Docker Agent capabilities to permission gates, recommending strict allowlists and scoped keys based on vendor claims and independent MCP research.
Infrastructure & RuntimeHosting Git on Cloudflare: What the Smart HTTP Protocol Actually Requires
Cloudflare Artifacts offers Git-native storage for Workers, but lacks documented consistency models and pack generation benchmarks, leaving critical protocol costs unverified.
