» Tag
llm
535 postsThe Real Divide in AI Verification: Code vs. Judgment
An AI agent's false 'I will remember' claim exposed why code-based gates, not judgment calls, are needed to verify agentic AI workflows.
LLM Latency Budgets: Speed Up AI Workflows Without Guesswork
A practical guide to building stage-level LLM latency budgets, workflow classes, and runtime enforcement for reliable AI products.
Study Finds 38.9% of AI Coding Agent PRs Contain Security Smells
Large-scale study of AI coding agent PRs finds 38.9% contain security smells, with most leaked credentials introduced by human developers, not AI.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comOpen-Source AI Incident Commander Hits 88% on 15 Real Outages
An open-source AI incident commander CLI passed 88% of checks across 15 real outages, running locally at near-zero cost with human approval gates.
Graphify Turns Codebases Into Knowledge Graphs, Cuts Tokens 71x
Graphify converts codebases into knowledge graphs, cutting per-query token usage by 71.5x. An open-source tool with 86k+ GitHub stars.
Qwen wrote its own app: why QuotePilot still needs a human approval gate
QuotePilot: an app Qwen models largely wrote and run, built around Decimal-based pricing and a mandatory human approval gate.
Fine-Tuning Qwen2-VL on AMD MI300X: The ROCm Pitfalls Docs Skip
A development log on fine-tuning Qwen2-VL for blockchain graph classification on AMD MI300X, detailing real ROCm setup and debugging friction.
How Airbnb Cut LLM Evaluation Time From Weeks to a Day
Airbnb's four-layer LLM evaluation stack uses deterministic caching and micro LoRA adapters to cut iteration time from weeks to a single day.
Qwisp: MoE expert-streaming engine runs Qwen3.6-35B-A3B on 8GB Macs
Qwisp streams MoE experts from flash to run the 35B-parameter Qwen3.6-A3B model on 8GB Macs, with bit-exact lossless decoding and raw-Metal speed.
Context Warp Drive: Deterministic, Zero-LLM Context Compaction
Context Warp Drive folds agent context deterministically with no LLM calls, cutting costs up to 72% while keeping 94% fact retention.