692 summarised stories in Tools & Coding, each linking back to the original source. Browse all topics →
Ben's Bites 4 weeks ago
Codex Record & Replay allows users to demonstrate a workflow once and converts it into a reusable, editable skill without manual coding. The feature can handle tasks like expense reports or time-off submissions, with benchmarks showing Sakana AI's Fugu achieving 73.7 on SWE-bench Pro. This enables non-technical users to automate recurring processes and reduces the need for custom development of repetitive task automation.
Together AI 1 month ago
ParallelKernelBench is a new benchmark for evaluating whether large language models can write optimized multi-GPU CUDA kernels, testing frontier models like GPT-5.5 and Gemini 3 Pro on 87 real-world problems from production codebases. The best model solved only 28 of 87 problems in zero-shot evaluation, with just 22 producing kernels faster than the PyTorch + NCCL baseline, and even with three sampling attempts the success rate peaked at 31%. Despite the poor overall performance, some generated kernels exceeded publicly available implementations, including a novel kernel for NVIDIA NeMo-RL's GRPO training loop, suggesting LLMs struggle with rank coordination and communication optimization but occasionally produce genuinely useful new code.
Hugging Face Blog 1 month ago
Transformers.js developers experimented with a proposed Cross-Origin Storage API to address redundant downloading and storage of shared AI model resources and WebAssembly files across different website origins. In a test case, the same 177 MB speech recognition model was downloaded twice when accessed from different origins, and a 4,733 kB WebAssembly runtime file was similarly duplicated despite being byte-for-byte identical. The Cross-Origin Storage API identifies files by cryptographic hash rather than URL or origin, allowing browsers to recognize and reuse the same resource across different websites, reducing bandwidth consumption and storage duplication.
Hugging Face Blog 1 month ago
Hugging Face automated its weekly release process for the huggingface_hub Python library using open-source tools, open-weights models, and human review for the release notes. The workflow reduced release time from 4–6 weeks to once per week, with a single GitHub Actions file handling version bumps, PyPI publishing, and changelog generation at approximately $0.25 per release. The system uses deterministic validation to ensure the AI-generated release notes include all merged PRs and prevents hallucinations by grounding the model with actual documentation diffs from each PR.
The daily briefing
TLDRocket reads 60+ sources, removes duplicate coverage, and summarises the day in two minutes. Free, no spam, unsubscribe anytime.
TLDRocket
22 July 2026
Cursor's launch of its Router—a request-level classifier that slashes AI coding costs by 30–50%—crystallizes the day's clearest economic story: the AI industry is moving from buildout to optimization. The system, trained on 600,000 live requests, routes routine tasks to cheaper models while reserving frontier reasoning for genuinely complex problems, exploiting a developer behavior pattern where 60% waste budget by defaulting to expensive models for everything. This mirrors Google's quarterly earnings, where cloud revenue surged 82% year-over-year to $24.8 billion on the back of enterprise AI adoption, proof that the company's staggering $180–190 billion annual infrastructure spend is finally yielding returns. Neither story is about capability breakthroughs; both are about efficiency and margin discipline.
Read the full briefing →