The latest Developer tools stories across our sources, prepared from the publishers’ own reporting.
Cloudflare Traces is now in open beta, automatically turning security rules, transforms, cache, routing and origin handling into OpenTelemetry spans without instrumentation. From December 1, tracing pricing moves to ingestion and retention volume, with 50GB ingestion and 10GB-month storage included on paid plans.
IEEE Educational Activities, with IEEE Computer Society support, launched a five-course program on AI processor architecture covering design fundamentals, advanced architectures, NPUs, emerging trends, and edge/cloud/quantum/IoT deployment, delivered partly through dialogues between AI-generated avatars.
LangChain's Managed Deep Agents v0.9 adds a reactions API that lets Slack agents set emoji responses dynamically, using a static string, a Python callable, or a decision model to build loading states and task acknowledgments.
Simon Willison released ttok 1.0, changing the CLI's default tokenizer from the older GPT-4 one to the GPT-5/GPT-6 tokenizer. The switch followed an experiment showing seven newer GPT models counted an identical 44,794 tokens across 31 fixtures.
Simon Willison released ttok 0.4, a maintenance update to his CLI token-counting tool built on OpenAI's tiktoken library. It fixes a Click warning, refreshes CI, and adds a --list-models command, with usage still available through uvx.
Databricks has launched a public Beta of a Workday Data Connect federation connector for Unity Catalog. The connector lets Unity Catalog access Workday data without moving or copying it.
Databricks introduced Lakebase alongside an agentic software development lifecycle concept, pairing databases with branching to support coding agents. The company says AI has changed how software gets built as coding agents take on a growing share of the work.
NVIDIA detailed six Omniverse projects in which developers directed frontier AI models such as GPT-6 Astra to build simulation applications, from warehouse humanoid control to browser-based ISS models. Tasks were driven by natural-language prompts, with Omniverse libraries supplying GPU-accelerated physics, rendering and sensor simulation.
NVIDIA detailed a five-step workflow for turning CAD files into simulation-ready OpenUSD assets for robotics, demonstrated on an ABB YuMi robot with assistance from a frontier AI model. The validated asset completed eight gripper pick-and-place cycles in Isaac Sim.
AWS published a reference architecture for sharing one Amazon SageMaker HyperPod EKS cluster across multiple teams using AWS IAM Identity Center, per-team SageMaker AI domains, Kubernetes namespaces, HyperPod Task Governance and namespace-level cost allocation.
Goodfire launched internal-activation monitors that watch an AI model's own computations instead of paying a second model to reread its output, available to Baseten customers. In tests on Kimi K3, monitoring about 1,500 sessions cost roughly $51 versus $233 for a cheaper external model and about $10,000 for a top-tier one.
IBM's torch-spyre team has made Spyre a native PyTorch device via PrivateUse1, giving tensors a real device identity on device="spyre" backed by PyTorch's allocator, streams and events. Compiled artifacts now launch as prepared recipes of typed operations instead of a second runtime graph.
LegalOn reduced its estimated daily Codex costs by 65% while maintaining development speed, according to OpenAI. The company matched Astra, Sol and Luna to tasks and managed budgets strategically.
A Hugging Face blog post describes using the ML-intern agent in HuggingChat to build six custom models in a few days, including a 0.8B prompt rewriter distilled from a 9B teacher and trained for about USD 16 in compute. Each project ran from a single prompt, with the agent handling training, evaluation and Hub publishing.
Developer Brandon Thomas launched Artcraft, a suite of seven open source apps recreating Adobe's Photoshop, Illustrator, Premiere, Lightroom, After Effects, InDesign and Acrobat Pro, built with Anthropic's Claude Opus 5.5 in Rust and released under MIT/Apache licenses.
LangChain added three capabilities to skills in Deep Agents: binding tools to skills, pinning skills at runtime, and reloading skills mid-thread. Tool schemas stay out of context until the agent reads a skill, and pinned skills skip a read round trip.
Amazon Quick and Amazon Bedrock Knowledge Bases now check document permissions with the authoritative source at query time, adding real-time ACL verification on top of existing pre-retrieval filtering. Mondelēz International has deployed Amazon Quick for more than 35,000 employees across four regions.
GitHub reports that one in three pull requests now involves an AI agent, up from fewer than one in 10 a year ago, and has built a fine-tuned classifier with Microsoft Applied Sciences that extends push protection to unstructured secrets, assessing candidates in under two milliseconds.
Databricks announced the general availability of building permission-aware Databricks Apps with on-behalf-of-user authorization. Databricks Apps lets developers build and deploy data and AI applications directly on the Databricks platform.
Microsoft Research Asia has open-sourced Agent Lightning v1.0, a roughly 3,500-line agentic RL framework that trains an agent using its real deployment harness via an OpenAI-compatible LLM proxy. Its coding example lifted Qwen3.5-9B from 41.8% to 56.4% Pass@1 on SWE-bench Verified with about 6,000 training samples.