#Deployment/Engineering

AIKrea (@krea_ai) posted a link to an agent app hosted on its platform. The post provides no further description of the app's features, capabilities, or purpose.

AIFactory's Automations, now generally available, let users describe a recurring workflow, set a schedule or event trigger, and have its Droid run it, with the model chosen per task. Templates cover ticket-to-PR, code review, security audits, PR babysitting, and morning Slack briefs. Among enterprise organizations using Automations in the past 30 days, 52% used automated code review, 48% used security review, and 35% used AutoWiki.

AINumerical differences between a rollout engine and a trainer can make reinforcement learning collapse even when algorithm and data stay identical. In a GLM 5.2 experiment, reward fell from about 0.9 to under 0.2 around step 20 without alignment, while aligned numerics kept reward stable over 25 steps. A Qwen3.5-MoE investigation traced a significant mismatch to how expert outputs were combined, and router replay alone was judged insufficient.

AIPromptArmor disclosed that a malicious Skill could hijack Copilot Cowork's AI gateway to spawn cloud agents that exfiltrate a victim's files to an attacker's server. No human approval was required, and any data Copilot could access was exposed. The vulnerability was reported to Microsoft on July 14, 2026, and Microsoft confirmed a fix on September 2, 2026.

AIGoogle Developers Blog describes how Sparse VideoGen (SVG) routes video diffusion attention heads into spatial or temporal sparse masks and implements them as custom JAX and Pallas Splash Attention kernels on TPU v6e. In isolated single-chip tests with 75.6K tokens and 10 heads, the sparse variants retain about 38.87% of query-key pairs. The article argues that theoretical sparsity must be converted into hardware tile skipping to yield real speedups.

AIOpenClaw (@openclaw) posted a link to a GitHub repository called openclaw-enterprise, tagging Red Hat, NVIDIA, and OpenAI. The post provides no further details about the repository's contents, purpose, or any partnership among the named companies.

AIThe vLLM project announced a strong presence at PyTorchCon North America, with core maintainer and Inferact CEO Simon Mo giving the keynote on scaling open frontier inference infrastructure. Other vLLM maintainers, including Nick Hill and Red Hat AI engineers, will lead a developer session and talks on agentic inference and attention.

AIToken traffic on OpenAI's Responses API has grown 100x compared with a year ago, while the team maintained 99.9% uptime and significantly improved performance. Sherwin Wu called scaling this quickly while preserving reliability hard to overstate, praising the year's work.

AIMicrosoft's Azure AI team argues that better models do not eliminate the need for a dedicated content extraction layer, since agents need trustworthy, structured, and auditable inputs. The post notes that building extraction directly on an LLM quickly demands chunking, layout parsing, grounding, normalization, and evaluation infrastructure. Microsoft positions Azure Document Intelligence and Azure Content Understanding in Foundry Tools as managed options for that layer.

AIHarrison Chase argues that as models get smarter, agent harnesses shift from deciding for the model to offering capabilities to it. In deepagents, each subagent gets its own model, tools, and permissions, while the main loop chooses which subagents to spawn and when.

AIOpenCode announces that GPT 6.1 Sol is now available within its platform. The post provides no further details on pricing, capabilities, or availability limits.

AIOpenAI announced that ChatGPT Plus and Pro subscribers can now use their subscription's AI usage across third-party tools. Notion confirms that users can connect their ChatGPT plan, select a GPT model, and use Notion's AI features.

AIOpenAI's Ultrafast mode for Astra runs up to 8x faster than Astra Standard and 4x faster than Astra Fast in Codex. Sherwin Wu says inference now feels nearly instant, shifting the bottleneck to users' next complaints.

AIBaseten announced a partnership with OpenAI that makes open models powered by Baseten available to OpenAI customers for multi-model agentic coding. The company says organizations can route each task to the best-fit open or closed model, with Codex and GPT models among the options, and that Baseten's day-zero access to new open models lets teams evaluate them quickly. Baseten also cites US-based infrastructure with zero data retention for all prompts and capacity across more than 90 clusters in 20+ clouds.

AICognition's post links to a blog announcing Sign in with ChatGPT, a login option for Devin. The post itself gives no further details about how the feature works.

AICognition says usage of Devin will count toward the Codex and ChatGPT Work usage already included in a user's plan, with an option to set a lower Devin limit. OpenAI models in Devin Cloud's Lite, Normal, Ultra, and Fusion modes will be paid for by the ChatGPT plan once the account is connected, while other models use the Devin quota.

AIKilo announces a blog post explaining how its sign-in with ChatGPT feature works. The main post provides only a link to the full explanation, so no further technical details can be confirmed from this source.

AIOpenAI says Ultrafast is available today for GPT-6 Astra in Codex, ChatGPT Work, and the API, with GPT-6.1 Sol coming soon. To access Ultrafast in Codex and ChatGPT Work, users need the new Pro 500 plan, which offers the highest usage limits at 25x Plus.

Why it matters: The post names the access surfaces and a new Pro 500 plan with 25x Plus usage limits, which matters for anyone deciding how to reach the Ultrafast mode.

AINoam Brown reported testing OpenAI's dots, powered by GPT-6 Astra, which identified recurring charges that could save him about $500 per year. He said the agent also connected to customer service and handled a full texting conversation on his behalf.

AIGoogle Gemma's account thanks the vLLM project and developer @mmastrac, linking to a GitHub pull request (vllm-project/vllm #57250). The post gives no details about the change itself.

AIGoogle Gemma says DiffusionGemma can be turned into a Jev-like model by seeding a canvas with a response template in vLLM. The approach extracts confidence scores and probability distributions for yes/no, multiple-choice, and scored questions in a single denoising step.

AICognition, the company behind Devin, shared a link inviting readers to try the product at The post provides no further details on features, pricing, or availability.

AISierra is a launch partner in OpenAI's new marketplace, letting eligible customers build Sierra agents using their existing OpenAI commitments. The company says almost 50% of the Fortune 50 work with it, and it positions its pricing around outcomes such as closed sales and retained customers rather than tokens.