#Engineering

Baseten collaborates with NVIDIA to ensure AI agent safety through OpenShell and Blaxel sandboxes

Baseten becomes a launch partner of the NVIDIA Agent Safety Platform, collaborating with NVIDIA OpenShell and Blaxel sandboxes to advance the open and secure frontiers of AI

How GitHub Copilot uses canvases to build interactive experiences

GitHub Copilot introduces canvases, allowing users to create customizable interfaces to interact with AI, replacing traditional chat interfaces; users can build Connect 4 games and Winget package manager UIs in Copilot using canvases, enabling two-way communication with AI and local code execution, improving efficiency and reducing token consumption.

LangChain fine-tunes on Baseten Loops via LangSmith tracing

LangChain announced the launch of the LangSmith fine-tuning feature at the Interrupt 26 conference in New York today. The feature is based on the open-source CLI smithtune, which reads LangSmith traces to generate supervised fine-tuning datasets and trains them on Baseten Loops, finally deploying them to Baseten's dedicated inference service. The CLI uses the user's Baseten API key, and training and inference costs are charged to the Baseten account.

GitHub Security Lab Taskflow Agent implements automated fuzzing

GitHub Security Lab Taskflow Agent automates the fuzzing process for C/C++ projects using LLM, completing entry point identification, build system analysis, test case generation, AFL++ execution, coverage analysis, and vulnerability report writing without manual intervention. The Agent uses the Claude Sonnet 5 model and manages state via an SQLite database to achieve iterative coverage improvement, doubling the time budget for each iteration until coverage growth stagnates.

Gemini 3.8 Live with Live Avatar released, integrating real-time visual presentation into conversational AI

Gemini 3.8 Live with Live Avatar released, enhancing near real-time visual presentation capabilities for Gemini conversational AI by natively coupling real-time video generation with speech, supporting 97 languages simultaneously, and available in Gemini Enterprise.

OpenHands Enterprise new features: enhanced visibility, governance, and control capabilities.

OpenHands Enterprise's latest version has added organizational-level features such as management credentials, usage, integration, and automation, and updated the interface aligned with Agent Canvas. Administrators can configure organizational key sharing, monitor usage through the 'Your Budget' dashboard, support OAuth authentication to connect to MCP servers, and achieve cross-organizational visibility and management automation.

How does GitHub Copilot application render large pull requests?

GitHub Copilot application addresses performance issues when rendering large pull requests with extensive comments by highly decomposing the document into two independent domains: deterministic code height and dynamic block effective height. The dynamic block geometry covers unpredictably tall content such as comment threads, drafts, and reply composers, each block identified by its content type rather than location and anchored to files, lines, and sides to ensure no loss during reordering. This design makes comment heights predictable before rendering, avoiding scroll jumps and enhancing rendering performance for large code differences.

Google DeepMind introduces Private AI Compute, achieving cross-device private persistent memory.

Google DeepMind introduces Private AI Compute, a new architecture that achieves cross-device AI persistent memory through cloud-based encrypted storage and device key management, ensuring data privacy. The architecture temporarily decrypts user data in a cloud-isolated environment to process requests and immediately encrypts and stores new contexts, while the device holds the decryption keys. Users can verify the authenticity of server software and audit security protocols, supporting AI in maintaining device-level privacy standards even when processing complex tasks efficiently in the cloud.

Release of Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS: Google DeepMind introduces the latest audio generation models.

Google DeepMind releases Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS, supporting the creation of custom character voices from scratch and direct control of scene dialogue.

NVIDIA Nemotron 3 Diarization: Real-time speech tag recognition costs only 0.01 dollars per minute.

NVIDIA Nemotron 3 Diarization model achieves real-time speech tag recognition with a Transformer architecture of about 100M parameters, with a cost of only 0.01 dollars per audio hour.

What is an AI dark factory? Software engineering insights learned from unmanned manufacturing.

AI Dark Factory refers to AI agents autonomously completing software development, with humans only setting goals and constraints. Its core principle originates from unmanned manufacturing, i.e., achieving automation through standardized inputs, deterministic verification, and continuous monitoring, rather than relying on model intelligence improvement. Engineering needs to refactor processes around automation, taking repeatable environments, executable acceptance criteria, and execution records as the foundation. Currently, it is mainly applicable to boundary-defined repetitive tasks.

Liquid AI collaborates with Qualcomm Technologies to bring personal AI context to Snapdragon processor-driven devices.

Liquid AI announced at the Snapdragon Summit 2026 that its device-side context layer, Liquid Context, has been optimized for Snapdragon® processors, specifically using Qualcomm® Hexagon™ NPU.

Cursor introduces Rollouts and Security Review two code deployment safety Bots

Cursor launches Rollouts and Security Review, two code deployment security Bots. Rollouts monitors the deployment process to report environment health status, and Security Review reports exploitable vulnerabilities in each pull request. Both are now available on Teams and Enterprise plans.

How to Self-Host AI Models: Complete Setup Guide

Self-hosted AI models allow organizations to directly control the reasoning runtime environment, deployed model weights, and service operation methods, suitable for scenarios involving sensitive data, pursuing cost-effectiveness under continuous utilization, model customization, and controlling the model lifecycle. By deploying open-source weight model checkpoints on local workstations, private servers, or controlled cloud infrastructure, and using reasoning runtime service models such as vLLM, Ollama, or llama.cpp, applications can call self-hosted endpoints just like calling managed model APIs.

What is model-agnostic in AI? A practical guide

Model-agnostic AI reduces direct dependency of applications and workflows on a single model, handling differences in authentication, request formats, model identifiers, etc. between different providers through an abstraction layer. OpenHands's model evaluation shows no single best model, its workflow supports use of different models for planning, implementation, debugging, and review, making model selection a continuous workload decision rather than a one-time vendor choice.

Step-5-Preview-BF16 model release by Jieyue Xingchen

Jieyue Xingchen releases flagship model Step-5-Preview-BF16, with 600B total parameters, 27B active parameters, supporting up to 1M context windows, excels in software engineering, professional knowledge work, and financial fields, capable of processing text, images, and video, deployable via OpenAI-compatible API or offline generation.

Jieyue Xingchen Step-5-Preview model release: Sparse MoE architecture with 600B parameters

Step-5-Preview, the flagship model from Jieyue Xingchen, adopts a sparse MoE architecture with 600B total parameters and 27B active parameters, supporting up to 1M tokens in context, and excels in software engineering, professional knowledge, and financial tasks, demonstrating outstanding performance in the DeepSWE v1.1 benchmark. The model can be deployed via OpenAI-compatible APIs using vLLM or SGLang, or used for offline generation with Transformers.

How to build an e-commerce shopping agent with Dify's new Agent.

Dify's new Agent can be used in e-commerce stores to help customers explore catalogs, understand product differences, and narrow down choices. It can also route high-value inquiries to team email. The initial version requires the Agent to first familiarize itself with the catalog, describe tasks through dialogue, search DOWSE's public product pages, and return product names, prices, brief descriptions, and links. After upgrading, the Agent can identify customer budgets, occasions, recipients, and preferences, recommend up to four verified products and explain the reasons, while also returning product images.

Baseten Labs releases Safety Infrastructure for Open Models paper

Baseten Labs releases Safety Infrastructure for Open Models paper, proposing to build a safety assessment and monitoring infrastructure for open models using Goodfire and Hugging Face.

Cognition partners with AWS to help businesses deliver products faster

Google releases EmbeddingGemma 2 multimodal embedding model

Google has released EmbeddingGemma 2, a multimodal embedding model with 740M parameters, supporting unified 768-dimensional vector representations for text, images, video, and audio. The model can run on consumer-grade hardware, providing low-latency semantic representations suitable for applications such as search, retrieval-augmented generation, etc. EmbeddingGemma 2 has performed excellently in multiple benchmark tests, such as text, code, visual document, and audio embedding benchmark tests, and supports vector truncation to reduce storage requirements.

How to build paid media agents with LangChain

LangChain publicly shares its construction method for paid media intelligence agents, which run on Slack, summarize ad platform and warehouse data weekly, and generate reports. Paid media increased from contributing 0 to 20% of the marketing pipeline within 6 months, and the cost per qualified lead decreased by 30% from June to August.

Devin Desktop & CLI introduces the Fusion architecture, enhancing code generation efficiency

Devin Desktop and CLI are now integrated with the Fusion architecture, improving efficiency by up to 39% compared to other code generation frameworks. Fusion allows users to select a cutting-edge model for planning and review ('lead') and a more cost-effective model for execution ('sidekick').

Cloudera collaborates with Mistral to bring dedicated, sovereign intelligence to enterprise data

Cloudera collaborates with Mistral to integrate Mistral models into Cloudera's hybrid data platform, enabling businesses to deploy AI models in private clouds, public clouds, on-premises, and fully isolated environments while maintaining full control; Mistral allows businesses to train AI models with proprietary data in a controlled environment, transforming decades of institutional data into custom AI models while retaining ownership of both data and intelligence.

Cursor launches Projects feature

Cursor launches the Projects feature today, supporting the handling of large workloads such as feature development, migration, or full applications, capable of maintaining context across months and automatically assigning tasks to thousands of sub-intelligent agents. The project runs independently in the cloud, synchronizing codebase knowledge and work preferences through shared files, coordinating agents responsible for planning and task allocation. The feature is currently in Beta and is gradually being rolled out to all users starting today.

Add automatic reminder function to Calendly using Dify Agent

Users create an Agent on the Dify platform to connect Calendly and Gmail tools, achieving the function of automatically sending daily meeting reminder emails. The Agent uses Calendly MCP to obtain meeting schedules and participant information, and refers to preset email templates to send reminders through Gmail without modifying Calendly itself. This solution is suitable for SaaS products that only need to supplement a single workflow function.