#Tutorial/How-to

AIMatei Zaharia argues AI tokens are now a resource to optimize in software engineering, with companies routing all AI usage through an AI Gateway. The approach enables centralized analysis, which found settings on Claude Code and Codex that can substantially lower cost, plus smart routing and per-task budgets for engineers.

AIDatabricks published an analysis of four techniques it used to reduce internal AI spend while growing adoption, with savings of up to 90% in some scenarios. The techniques are shifting defaults to cheaper models such as GLM, automated task-level model routing, per-user spend visibility with adaptive budgeting, and pruning context bloat. The author, Ali Ghodsi, reposted Databricks co-founder Patrick Wendell's summary and recommended it.

AIHugging Face's broadcast explains how AI agents were used to reproduce papers accepted at ICML 2026. The post itself gives no further details on methods, results, or success rates.

AIMckay Wrigley recommends putting "You are AGI-pilled." in the system prompt for all agents, claiming it produces a much better experience. He argues that current agents behave as if the world will stay static, and that this line makes them talk and act more like AGIs.

AIMeituan LongCat invites users to try its agentic coding demo videos, each of which includes its prompt, using LongCat-2.0. The post directs readers to the LongCat API platform and a Discord community to share their own creations.

AIOpenRouter has released Ori Eval, an agent-driven tool that runs your app's prompts against candidate models and returns a comparison table of catch rate, latency, cost per PR, and pass/fail results. The tool asserts on called tools and grades open-ended answers with an LLM judge, pinning the harness and model during each run. Its evals are code files that can run in CI to block regressions and re-run when new models ship.

AIFireworks AI ran controlled SFT experiments on Qwen3.5-9B comparing LoRA with full parameter fine-tuning across three synthetic verifiable tasks. The post argues that a FullFT advantage can come from data coverage, learning-rate tuning, or adapter rank, and it recommends testing these in that order before switching methods. Under a fixed multi-task budget, FullFT kept a 4.29-point lead over the best LoRA recipe tested, while matched data exposure favored LoRA.

AILiquid AI doubled the LFM2 tokenizer's vocabulary from 65K to 128K without retraining from scratch, extending the original BPE merges and initializing new embeddings as the mean of their sub-tokens. The expanded tokenizer needs 4.0× fewer tokens for Thai, 2.6× fewer for Vietnamese, and 2.4× fewer for Hindi, which the source says yields roughly 2.2–3.7× faster on-device decoding for these languages with no reported quality loss on previously supported languages. LFM2.5-8B-A1B and the expanded tokenizer are available on Hugging Face with open weights.

AIFireworks AI describes fine-tuning Qwen3-Embedding-8B on private (query, positive) pairs using bidirectional InfoNCE loss through its Training SDK, then serving the model via an OpenAI-compatible embeddings endpoint. The post reports that around 150 training steps was enough, that rank-32 LoRA landed within about one point of full-parameter fine-tuning, and that gains were largest where the base model struggled, while tasks like CoSQA and FiQA2018 showed flat results.

AIHugging Face announced a broadcast titled "Training Agents 3" that teaches viewers how to train a local, open-weight agent using reinforcement learning. The post provides no further details on models, tools, or benchmarks.

AIFireworks AI has made Kimi K3 available for Multi-LoRA serving and training in private preview through Fireworks Serverless Training. The post explains how small LoRA adapters can be trained on K3 and served with live merge or multi-LoRA deployment, and it reports two example tasks, Countdown and Frozen Lake, with reward curves.

AITrilogy released an AI Cybersecurity Playbook built around Kimi K3, served on Fireworks through an OpenAI-compatible API. The playbook pairs deterministic tools for indexing, parsing, and deduplication with Kimi K3 for reachability, impact, and remediation judgment, followed by a separate verification stage.

AIAnthropic co-founder Mike Krieger said he had Opus 5 build a renderer, savegame parser, and boat, plane, truck, and bus simulator overnight. The project lets him walk around the world of Transport Tycoon, an idea he had wanted since he was six years old.

AIEthan Mollick's guide says ChatGPT and Claude are the main choices for real work, since their agent modes can act on a computer. He separates agent modes that run on the company's computers from those that access the user's own computer. He recommends keeping approval settings on for sending, spending, or deleting, because of prompt injection risk. He also notes that Gemini currently lags for agentic work, though its Notebook and video tools are useful.

AISkywork (@Skywork_ai) posted a story created with Skywork Video and linked to the product's sign-up page. A quoted post from @ivnways says the tool gives users more control over AI video output than typing a prompt and hoping for a useful result.

AIAndrej Karpathy describes using /voice to ramble for about 10 minutes, sometimes as a short interview, to give an LLM context that would be tedious to type. He says LLMs reconstruct these messy streams of thought remarkably well, often returning a cleaner version than the speaker started with, which improves shared understanding and reduces later corrections.

AICat Wu says she uses Claude Cowork to manage her calendar and shares her prompt. The prompt caps meetings at under 20 hours per week, excludes dinners from that cap, dedupes conflicting meetings, and learns from past declines via a refined skill, asking before updating invites.

AIThe article explains how reasoning models can offer multiple effort modes, separating training-time methods from inference-time controls such as system prompts and chat templates. It compares six open-weight models, including DeepSeek V4, Nemotron 3 Ultra, Kimi K2.5, GLM-5, Qwen3, and Inkling, noting that their reports disclose different levels of detail. It also shows how GPT-5.6's model selection and effort settings act as two separate scaling axes.

AIStability AI audio researcher CJ Carr will cover running the model locally, even on a MacBook Pro, in a workshop. Topics also include training your own LoRA, using Stable Audio with Ableton, and building custom audio tools on top of the model.

AIStability AI is addressing producers, creative coders, and curious readers who want to understand how its audio technology works under the hood. The post says basic comfort with music tools and Python helps but is not required.

AIAnthropic will host a live walkthrough at 10am PT tomorrow showing how Claude Code progressed to multiplayer Claude Tag. The session will cover how Claude Tag works, including monitoring channels, doing proactive work, letting the whole team steer it, and remembering prior instructions.

AICat Wu, owner of the account for Anthropic, shares a retrospective on making Claude Code with the product's early team. Anthropic's linked feature collects accounts from the builders and early users who shaped the tool.

AIAnthropic's Cat Wu asks users to share their top Claude Code use cases, workflows, and artifacts. She highlights a candidate-sourcing workflow in which Claude Code runs a dynamic workflow to find 100 candidates with LinkedIn, Twitter, blog, and podcast links plus a one-line pitch each, then emails the results as an artifact.

AISebastian Raschka's tutorial shows how to build a fully local coding agent by pairing an open-weight LLM served through an inference runtime with an open-source harness that can read files, edit code, and run commands. He recommends Qwen-Code for Qwen3.6, citing Nvidia's Polar paper, which found Qwen models performed best in Qwen-Code. The Qwen3.6 35B-A3B model is about 22 GB to download and needs roughly 30–40 GB of RAM.

AILilian Weng published a long blog post on scaling laws, which help estimate the best split of compute between data and model size before a large training run. The post covers what scaling laws predict, how compute-optimal allocation works, and why Kaplan et al. and Chinchilla reach different conclusions. It also addresses how data limits and fitting details make extrapolation difficult.