Engineering Insights

Chat SDK adds Dial support

Chat SDK now supports Dial with the new .vendor-official adapter Build bots that send and receive SMS, MMS, and iMessage on a real phone number, with bidirectional media and inbound voice-call transcripts. Replies use the standard Chat SDK thread and message APIs, with HMAC-verified webhooks and stable per-conversation threading. Each phone conversation becomes a Chat SDK . Texts and media arrive as messages on that thread, and when a voice call ends, its transcript appears there too. Subscripti

Chat SDK adds Photon support

Chat SDK now supports Photon with the new .vendor-official adapter Build bots that send and receive iMessages directly and in group chats, share media, and react with native tapbacks. Run the adapter against Spectrum Cloud, your own server, or directly on a Mac. Each Photon conversation becomes a Chat SDK thread, and tapbacks come through as reactions. Handlers, posts, and subscriptions work the same as with any other adapter. Webhooks are HMAC-verified, and your bot can reply to a DM straight f

The future of governing AI agents

Autonomous security agents need governance built into the architecture across reasoning separation, continuous evaluation, progressive trust, and agent-specific telemetry at scale.

Deadlocks and downtime

Every second* of every day, your Postgres database could be hunting with intent to kill. *or whatever your deadlock_timeout setting is. Deadlocks happen when multiple transactions each hold a lock that the other needs, blocking both from proceeding. When a transaction has been waiting on a lock for longer than deadlock_timeout, Postgres runs its deadlock detector. If it finds a cycle, it cancels one transaction in the cycle — freeing the others to proceed. Without this process, new transactions

Introducing GPT-Live

A new generation of voice models for natural human-AI interaction, now powering ChatGPT Voice.

Native-speed vLLM transformers modeling backend

AI Search - Filter AI Search list items by exact object key

In AI Search, you can upload files to an instance, or connect a data source such as an R2 bucket, to make your content searchable with natural language. Each file becomes an item identified by an object key (its filename or path). The list items endpoint returns the items in an instance. That endpoint now accepts a key query parameter, so you can look up a single item by its exact object key without paging through the full list. This complements the existing item_id filter for when you know the

Workers AI, AI Search - Workers AI toMarkdown and AI Search now supports GIF and BMP image conversion

Workers AI Markdown conversion (toMarkdown) now supports .gif and .bmp image files, in addition to the JPEG, PNG, WebP, and SVG formats already supported. GIF and BMP files run through the same image pipeline as other formats. Each image is resized if needed (and for animated GIFs, only the first frame is used), then passed to an object-detection model to identify what it contains. Those detected objects prompt a vision model that writes a natural-language description of the image, which becomes