About AI Coding Tools

AI Coding is the subset of AI focused on code generation, autonomous coding agents, and AI-enhanced developer workflows. OrangeBot.AI tracks releases of Claude Code, Cursor, GitHub Copilot, Aider, OpenHands, Codeium, and the broader agent-harness ecosystem. Also covers benchmarks, evals, and pull-request automation tools.

TOPIC · AI-CODING

AI Coding Tools

Copilot, Cursor, Claude Code, Aider, Windsurf, Devin, and the rest of the agentic-coding ecosystem.

62 unique stories from the last 14 days across 8 sources.

Hacker News(4)

  1. Claude Code and Codex reverse engineered my 1994 Jeep's ECU over a $3 Arduino [video] (www.youtube.com)
  2. Nintendo Takes Down Emulators and Firmware in Massive GitHub DMCA Sweep (www.ign.com)
  3. Goldman Sachs is pulling Claude Code from its bankers and traders (www.thestateofai.com)
  4. Claude Code sends your email address to websites (github.com)

GitHub Trending(2)

  1. cursor / plugins
  2. zed-industries / zed

Product Hunt(8)

  1. Stringman

    Turns your room into a crane game

  2. ROOTKey Platform

    See what changed in your data. Restore what you can prove.

  3. Talyn

    Run Claude Code and Codex in parallel cloud sandboxes

  4. ScroogeFrog AI

    AI ad fraud detection that protects your marketing budget

  5. AXL

    GTM copilot for AI product founders: 1,200+ actions via MCP

  6. Agent Clan

    Command your Claude Code and Codex agents with gestures

  7. Maritime

    Cloud computers for any AI agent, customized your way

  8. CC Auto Review

    macOS app that reviews GitHub PRs w/ your local Claude Code

Hugging Face(36)

  1. Weight-Space Programming Doesn't Need Transformers

    Standard neural architectures such as the Transformer were chosen because they train well with gradient descent and run fast on GPUs. In weight-space programming, a large language model (LLM) writes target weights by hand without gradient-based training, so those reasons no longer apply. Should we keep the Transformer? Given minimal guidance, ten labeled examples, and a sandbox, mainly on multiple-choice questions, we task three frontier LLMs with either generating the weights of a fixed architecture or designing the architecture themselves. When free to design, all three compile plain-text facts into a sparse table queried by a fixed procedure, reaching up to 54.4% test accuracy. An LLM-designed model achieves higher test accuracy than the most accurate released Transformer that the same LLM wrote while using about 1,800x fewer parameters. Without gradients, they choose a table, in which the LLM can state what each weight means and can find and rewrite a wrong one. Each wrong answer points to one row, and removing a row changes about one other answer. From this, we propose an associative-memory-based model up to 17x faster on GPUs than a same-sized Transformer. We also test pretrained weights, other benchmarks, and tasks where models generate answers. Pretrained LLMs remain more accurate; the claim concerns which structure to build when writing weights.

  2. Multi-Branch Transformer

    Transformer language models are commonly scaled by making their blocks wider or stacking them deeper. In this work, we explore a complementary path: branching, the parallel composition of full Transformer blocks. We introduce the Multi-Branch Transformer (MBT), which processes the input through shared initial layers, executes independently parameterized branches of sequential blocks in parallel, and fuses their outputs before shared final layers. At comparable training compute, we study three MBT variants: Wide-Shallow preserves block width, Narrow-Deep preserves sequential depth, and Balanced interpolates between them. In controlled pretraining experiments at four mixture-of-experts scales (0.8B to 2.1B total parameters), 2-branch Wide-Shallow MBTs consistently outperform sequential Transformers under matched parameter and training-compute budgets. Balanced MBTs improve further and retain comparable quality as the number of branches increases. To improve inference efficiency at long contexts, we introduce Joint-KV, which constructs a shared KV cache from branch representations at each depth. Our final recipe combines Balanced MBTs with Joint-KV, supported by an inference engine that exploits branch grouping and shared-cache reuse. Fitted scaling curves indicate that standard Transformers require up to 1.26x as much training body compute to match our final models' validation loss. These models reduce full-model KV-cache storage by up to 34% and deliver up to 51.4% higher decoding throughput on a single H200 GPU. Ablation studies show that branching yields gains beyond changes in width and depth, enables an effective KV-sharing strategy (Joint-KV), remains robust across fusion functions, and performs best in the middle layers. These findings show that branching offers a practical path for scaling language models toward both better modeling quality and faster inference.

  3. Breath-Dx: A Time Series Classification Benchmark for Disease Screening with eNose

    Electronic nose (eNose) technology offers a portable, non-invasive route for disease screening via exhaled breath analysis. However, progress in the field is hindered by the lack of systematic benchmarks that compare machine learning methods under realistic conditions, particularly sensor drift, a known but rarely quantified failure mode. Most existing studies rely on small, private datasets with inconsistent protocols, rendering results non-comparable and slowing progress. We introduce Breath-Dx, the first comprehensive benchmark of time series classification methods for breath-based disease screening. Our work leverages the Scent of Health (S-OH) dataset, the largest public clinical eNose collection to date, comprising 1,234 patients across nine diagnostic groups, each providing a 17-channel multivariate time series with explicit temporal train/test splits that account for sensor drift. We systematically evaluate 10 methods spanning four families: feature-based tabular learners, convolution-based models, graph-based, and time series architectures. All methods degrade significantly under temporal validation (0.05-0.12 AUC drop compared to random cross-validation), confirming that drift embeds a strong temporal signal. Under this realistic evaluation, the best methods achieve ROC AUC of 0.74-0.76 for some lung conditions, though others remain near chance. Recurrent architectures show the most consistent performance across diseases, while specialized time series methods achieve the highest peak results. Demographic confounders matter: breath signals provide independent diagnostic value for some conditions but add little beyond age for others. We release the S-OH dataset, all temporal splits, evaluation protocols, and code to enable reproducible research and accelerate progress in clinical breathomics.

  4. Self-Improvement Memory Enables Continual Recursive Self-Improvement

    Harness optimization is often studied with a fixed policy model and benchmark. When either changes, the optimizer must decide which parts of an evolved harness to reuse or revise, but the harness alone may not explain why earlier edits were made or how they performed. In this work, we study Continual Recursive Self-Improvement (CRSI), a setting for continuing existing harness-optimization methods across compatible benchmarks or policy models. We propose Self-Improvement Memory (SIM) for the harness optimizer, rather than the task-solving policy. SIM groups edits by recurring harness problem and records their rationale, predicted and observed outcomes, and evaluation context. New attempts extend this history as optimization continues. In our experiments across four domains, SIM achieves higher final scores than harness-only and root-cause-only continuation in cross-dataset and pairwise cross-model comparisons. Across ten independent source runs in coding, paired continuations with SIM improve final scores by an average of 2.36 points over harness-only continuation (p=0.003).

  5. Sage S2.1 Flash Technical Report

    We introduce Sage S1.5 Flash, a 320-billion-parameter general-purpose mixture-of-experts model designed to reason over scientific evidence and act in interactive environments. Sage Sci-Perception uses dedicated encoders for molecular graphs and biological sequences and represents spectra as ordered patches of measurements. Modality-specific projectors map these representations into the input space of a shared backbone, which processes them alongside text and images. To develop the model's agent capabilities, we construct verifiable environments for tool use, search, and coding, as well as specialized workflows involving machine learning experiments and scientific software. Task completion is checked against execution results, changes in environment state, or submitted artifacts. We evaluate Sage S1.5 Flash on 33 benchmarks across seven domains. In the primary comparison against five baselines, it ranks first on OfficeQA Pro (72.2), MLE-bench Lite (90.3 HumanRank), Harvey LAB, and SpreadsheetBench~2, and second on OSWorld-Verified (85.2), WebArena-Verified (67.6), and DeepResearch Bench~II (54.2). These results show particular strengths in using documentary evidence, producing verifiable workbooks, and carrying out machine learning tasks, alongside competitive performance in desktop and browser workflows that require the model to act on visual feedback.

  6. AllocFlow: End-to-End Portfolio Trading Action Generation from Raw Market Data

    Real-world portfolio management requires converting noisy market observations into coordinated, simultaneous allocations across multiple assets under shared capital constraints. Advances in forecasting models and alpha discovery help identify opportunities. However, traditional pipelines often decouple signal forecasting from portfolio construction, potentially misaligning learning with allocation and execution objectives. We present AllocFlow, an end-to-end flow-matching framework that directly generates joint portfolio allocations from raw multi-asset market histories. To capture dependencies among portfolio positions, AllocFlow employs a cross-asset flow Transformer conditioned on temporal market representations to generate continuous, signed position weights across all instruments jointly. To address the sample inefficiency and instability of learning continuous allocation policies from scratch, we establish a two-stage training scheme: strategy-supervised imitation shapes a structured distribution of candidate portfolios, followed by group-relative reinforcement learning (RL) that refines joint decisions using portfolio-level financial rewards and a frozen policy reference. On hourly cryptocurrency perpetual futures, RL refinement raises the annualized net return of the imitation policy from 27.79% to 51.38% and its Calmar ratio from 0.955 to 4.043, surpassing the teacher strategy on every metric, and AllocFlow reaches a Calmar ratio of 4.90 on six further months of unseen 2026 data without retraining.

  7. Envisioning Agentic Robotics: A New Era of Robotic Systems

    The release of GPT-6 Astra marks the emergence of a new paradigm in robotics: agentic robotics. Within weeks, frontier models with agentic harnesses matched or exceeded task-specific policies on many embodied benchmarks. Instead of training a dedicated policy for each family of tasks, an agentic robotic system applies a foundation model to open-world embodied tasks. It operates in a closed loop: observing the environment, interpreting user intent, reasoning about the task, invoking skills and tools, verifying outcomes, and adapting its strategy. Despite these advances, the community has yet to converge on a shared definition, a reliable evaluation framework, or a research roadmap. We argue that Embodied System 2, realized by a foundation model together with an embodied harness, forms the decision-making core of agentic robotics. Its central challenge is to translate general-purpose reasoning into effective and verifiable actions under physical constraints such as limited reaction time, mechanical limits, unpredictable environment changes, and irreversible consequences. Motivated by this perspective, we propose an evaluation framework that measures both model capabilities and closed-loop system performance. Furthermore, we identify the principles that carry over from digital to physical agents and the challenges that must be addressed to achieve reliable physical autonomy. On top of the Embodied System 2 and the evaluation framework, we envision a research roadmap toward sustained autonomy: the ability to perform work safely and reliably with progressively broader responsibility, from executing individual primitives to fulfilling standing roles.

  8. VibePrevisBench: Evaluating Agentic Coding for 4D Scene Generation in Blender

    Controllable real-world video generation models can condition on a rendered graybox proxy video, which fixes layout, motion and camera. Coding agents can build such videos by reconstructing a 4D scene in Blender, yet no benchmark checks both these videos and the long, multi-shot scenes behind them. We introduce VibePrevisBench, a benchmark for 4D scene reconstruction: an agent turns a brief into an executable 4D scene and renders its proxy video. Its 242 tasks span four input kinds, including scripts, photos, continuations, and reference videos. Each delivery is scored along two evaluation dimensions. Rule-based Scene Check re-runs the code and compares the exported 4D scene with hidden ground truth facts in nine dimensions. VLM-based Screen Check asks a VLM judge whether the facts that pass can be recognized in the proxy video. We analyze the evaluation results and trajectories of 11 frontier coding agents and identify three key observations. First, multimodal understanding separates the agents, especially in understanding space and camera motion. Second, all agents seem to already master basic camera work and staging, and fail at challenging requests. Third, agents inspect their own renders but fail to correct their errors. Agents thus need better visual understanding of space and camera, explicit checks for challenging requests, and item-by-item verification of renders against the brief.

  9. Rethinking in Spikes: Mitigating Hallucinations in MDLMs with Step-Aware Decoding

    Recent advancements in multimodal diffusion language models (MDLMs) have exhibited strong global modeling capabilities through iterative refinement. We observe that mask uncertainty follows an overall downward trend during iterative denoising, while a few denoising steps exhibit localized entropy spikes closely associated with hallucinations. We argue that anomalous steps can be identified by measuring the entropy deviation of each denoising step from its neighboring temporal window. We hypothesize that discrete token commitment discards the distributional information at masked positions, limiting the continued interaction between candidate semantics and the context during entropy spikes. With this goal, we present StepRefine, a plug-and-play decoding strategy that leverages semantic context to achieve reliable denoising. The core of our method lies in a step-aware switch between discrete denoising and latent refinement. The model employs probability-weighted embeddings at entropy spike steps to preserve diverse reasoning hypotheses and performs multiple rounds of continuous refinement. Meanwhile, a refinement controller monitors distributional stability to reduce unnecessary loops and switch the model back to standard discrete denoising. Through extensive experiments, StepRefine demonstrates significant hallucination mitigation across different MDLMs on multimodal benchmarks.

  10. G-CAST: Governed Continual Self-Training for E-Commerce Agents under Evolving Demands

    With the rapid advancement of Large Language Models (LLMs), an increasing number of users are turning to LLM-powered agents for conversational shopping in e-commerce. However, the user needs continue to change as new product categories, procurement scenarios, and feature requests emerge. At the same time, the product's execution harness-including its tools, skills, and operating procedures-continues to evolve to support increasingly sophisticated interactions. These changes create two practical challenges: 1) Demand drift alters the distribution, intent, and complexity of user requests, causing a model trained once to fall behind; 2) Harness evolution requires the model to use newly introduced capabilities correctly without losing behavior that users already rely on. Periodic manual retraining is too delayed and labor-intensive for this setting. To address these challenges, we present G-CAST, a governed continual self-training system that converts daily production evidence under evolving demand and capability contracts into reward-routed, entropy-prioritized training data and governed reward evolution. Governed checkpoint transition under cumulative evaluation controls each model update, enabling the agent to learn newly introduced behaviors without sacrificing previously accepted capabilities. Experiments on a real production e-commerce agent show that G-CAST achieves stronger adaptation to newly introduced demand while better preserving previously accepted behavior compared with a strong autonomous post-training baseline. Ablations further show that both reward-routed, entropy-prioritized data orchestration and governed reward evolution are important to these gains. G-CAST is now deployed in Yuancai(http://cai.1688.com/), a B2B procurement agent on Alibaba's 1688 platform.

  11. PhysLoop: Harnessing LLMs' Physics Awareness for Self-Improving Video Generation Models

    Video generation models produce visually realistic videos but still struggle with physical dynamics and causal consistency. Learning from generated videos requires fine-grained physical supervision, yet reliable quality judgments, correction pairs, and controlled variations are difficult to obtain. Large language models (LLMs) possess physical knowledge, but struggle to reliably guide video generation models in producing their training videos. We present PhysLoop, a unified harness designed to support physically grounded self-improvement in video generation models. The harness guides and constrains LLMs throughout video generation, selection, and correction, distilling their physical knowledge into physically consistent videos. The framework supports controllable training-data construction and inference-time video repair. By identifying and correcting failures, PhysLoop marks a key milestone toward recursive self-improvement (RSI) in video generation models with limited human intervention. We release 100K videos spanning diverse, realistic indoor and outdoor scenes, along with generation tools and two evaluation datasets: PhysLoop-Gen and PhysLoop-Repair. We evaluate controllable generation and localized repair on these datasets, respectively, and assess physical behavior on the external Physics-IQ Verified benchmark.

  12. Duo: A Reciprocal World Model for Full-Duplex Interaction

    In this report, we introduce the Reciprocal World Model (RWM) and present Duo v1, a full-duplex world model that continuously perceives audiovisual inputs while responding to users in real time. The RWM treats intelligence and physical reality as equally fundamental to what constitutes a world. Existing world models largely focus on simulating the latter, leaving reasoning and decision-making as external controls rather than intrinsic parts of the world itself. In RWM, we identify two equally fundamental capabilities: Intent, which provides the cognition underlying world interactions, and Realization, which generates coherent audio-visual behavior. To realize this formulation, Duo v1 introduces three key technical designs: (1) Leveraging autoregressive modeling for continuous interaction reasoning and participation decisions, rather than restricting it to instruction understanding for diffusion models. (2) Constructing chunk-specific conditions and synchronize tokens across Intent and Realization on a unified timeline. Module-specific attention masks enforce causal dependencies both within and across chunks, preventing access to future inputs. (3) Mitigating distribution shift in both modules through dual rollout distillation, training on autoregressively generated Intent tokens and synthesized audiovisual histories from Realization. Both modules continuously perceive real-time audio, video, and text, enabling full-duplex interaction with spontaneous conversations and interruptions. Beyond conversation, users can share real objects, reshape virtual scenes, guide assistants, or play audiovisual games without mouse or keyboard controls. The release of Duo v1 makes our first step, with the next version coming soon. We look forward to broader exploration of this emerging field.

Techmeme(8)

  1. Sources: Bret Taylor's Sierra is in talks to raise ~$1.5B at a ~$30B valuation, up from $15.8B in May; it said in October it hit $400M+ in annualized revenue (The Information)

    The Information : Sources: Bret Taylor's Sierra is in talks to raise ~$1.5B at a ~$30B valuation, up from $15.8B in May; it said in October it hit $400M+ in annualized revenue —  Sierra, the AI customer support startup founded by former Salesforce co-CEO Bret Taylor, has been in talks to raise new funding …

  2. Nvidia reports Q3 revenue up 103% YoY to $115.8B, above $109.1B est., Data Center revenue up 110% to $107.6B, and forecasts Q4 revenue above estimates (Nvidia Newsroom)

    Nvidia Newsroom : Nvidia reports Q3 revenue up 103% YoY to $115.8B, above $109.1B est., Data Center revenue up 110% to $107.6B, and forecasts Q4 revenue above estimates —  - Revenue of $115.8 billion, up 103% from a year ago  — Data Center revenue of $107.6 billion, up 110% from a year ago

  3. Sources: Jensen Huang has discussed a comeback as Nvidia CEO with confidants, as allies form a "war room" to lobby for his return and seek White House support (Financial Times)

    Financial Times : Sources: Jensen Huang has discussed a comeback as Nvidia CEO with confidants, as allies form a “war room” to lobby for his return and seek White House support —  Tech supporters lobby chipmaker's board to reinstate former CEO after he resigned following affair with alleged Chinese spy

  4. Sources: Nvidia appealed to Taiwanese officials after its manager was charged over chip smuggling, and sought advice from TSMC, which declined to get involved (Mackenzie Hawkins/Bloomberg)

    Mackenzie Hawkins / Bloomberg : Sources: Nvidia appealed to Taiwanese officials after its manager was charged over chip smuggling, and sought advice from TSMC, which declined to get involved —  Nvidia Corp. appealed to Taiwanese officials after prosecutors on the island accused an employee of helping to smuggle AI chips into China …

  5. Sources: Anthropic's IPO is about six times oversubscribed, and bankers have told prospective investors that existing backers will be prioritized (Bailey Lipschultz/Bloomberg)

    Bailey Lipschultz / Bloomberg : Sources: Anthropic's IPO is about six times oversubscribed, and bankers have told prospective investors that existing backers will be prioritized —  Anthropic PBC's initial public offering has drawn about six times the orders for the shares available, according to people familiar with the matter …

  6. OKX and ICE's joint venture OKXICE plans to launch a US tokenized stock venue on November 23 and list 70+ assets; 56 of the 62 companies it targeted objected (Scott Patterson/Bloomberg)

    Scott Patterson / Bloomberg : OKX and ICE's joint venture OKXICE plans to launch a US tokenized stock venue on November 23 and list 70+ assets; 56 of the 62 companies it targeted objected —  Crypto's race to grab a slice of the US stock market is picking up speed even as many of the companies whose shares are being targeted for tokenized trading are pushing back.

  7. Anthropic says during prerelease testing Mythos 5.2 and Fable 5.2 formalized the proof of Fermat's Last Theorem and solved two open black hole physics problems (Jay Peters/The Verge)

    Jay Peters / The Verge : Anthropic says during prerelease testing Mythos 5.2 and Fable 5.2 formalized the proof of Fermat's Last Theorem and solved two open black hole physics problems —  Claude Fable 5.2 is designed to take on big technical challenges, and it's available today.

  8. Exponential Time Hypothesis co-proposer Russell Impagliazzo calls OpenAI's disproof "shocking news"; it could one day speed up chip design and break encryption (Cade Metz/New York Times)

    Cade Metz / New York Times : Exponential Time Hypothesis co-proposer Russell Impagliazzo calls OpenAI's disproof “shocking news”; it could one day speed up chip design and break encryption —  A longstanding hypothesis held that some problems would always be hard for computers to solve quickly.

Solidot(4)

  1. Windows 收集用户计算机安装的编程智能体和开发工具

    一位开发者利用 AI 工具寻找 Windows 未记录的功能,结果发现了一个 DLL 文件记录了用户使用的编程智能体如 codex.exe、claude.exe、copilot.exe 和 cursor.exe 等,以及开发者工具如 python.exe、dotnet.exe、 gcc.exe 和 code.exe 等。该 DLL 文件是 MissionControlAggregator.dll,是一个运行中的 Windows 组件,它会将记录的信息保存在本地,报告给 Windows 诊断系统。除非用户选择退出诊断数据收集(默认启用),那么这些数据最终会发送给微软。MissionControlAggregator.dll 通过字符串比较操作检查一系列 AI 智能体和开发者工具。

  2. 微软云服务宕机持续逾 24 小时

    微软周六表示,影响各国企业和政府所用 Microsoft 365 应用的全球服务中断已经结束,这次中断持续了超过 24 小时。微软在 X 上发文说:“我们已完成解决这一事件根本问题的工作,并为其余受影响的客户和地区恢复了服务。”Teams、Outlook 和 Copilot 等微软应用都受到了这次服务中断的影响,导致数十个国家出现通信问题。微软表示,“没有发现”任何可能导致这次服务中断的外部干扰,并将问题归咎于服务器故障。这次中断是微软历史上持续时间最长的中断之一,也是今年的第二次。英国国家网络安全中心前负责人 Ciaran Martin 认为这次宕机事故在规模上类似亚马逊 AWS 和 CrowdStrike 事故,其影响可能更为严重。日本政府的 IT 基础设施也受到影响,英国国家医疗服务体系(NHS)工作人员不得不改用纸质记录,无法调用线上存储的资料。这次中断凸显出,各国政府和企业普遍依赖极少数大型科技公司,而这种依赖充满风险。

  3. 微软云服务全球宕机

    微软云服务全球宕机,受影响服务包括了 Outlook、Teams 和 OneDrive 等办公工具以及 AI 工具 Copilot。微软通过其社交媒体账号表示正全力解决该问题。根据 Downdetector 的报告,截至周四 18:00 GMT,全球报告的 Outlook 故障数量接近 14 万。目前无迹象表明此次大范围宕机是由网络攻击引起的。微软表示,受影响的用户可能无法查看部分文件,或无法在日历应用中查看详细信息。Copilot 也可能无法正常运行。即时通讯和在线会议功能基本保持正常。微软称此次事故是由部分服务器意外不可用导致的,这给其余系统造成了沉重负担。与微软有合作的 AI 巨头 Anthropic 和 OpenAI 也报告了故障。

  4. 高盛限制交易员和银行家使用第三方 AI 智能体

    高盛限制交易员和银行家使用 Anthropic 的 Claude Code 和微软的 GitHub,此举主要是出于风险方面的考虑。AI 编程工具主要提供给软件工程师,以及负责交易、量化研究和软件工程等的策略师。交易员将无法再使用这些工具监控客户交易流、模拟市场走势和持仓,以及制作文字和图表。此次限制适用于能独立编写和运行代码的第三方智能体。以 Claude Code 为例,它能读取文件、编写程序并运行,通常拥有与使用者相同的权限。工程师用此类工具编写的代码通常会经过审查、测试和部署控制,然后才会投入使用。交易员编写的脚本则可能未经任何审查就被投入日常使用。使用智能体模拟仓位的交易员不太可能了解代码如何处理异常情况,在交易台上这可能意味着损失。

Browse other topics