Summary of the Keras community call — 1. New pluggable backend plan, upcoming MLX and PaddlePaddle backends; new backends are installed as add-on libraries 2. Reducing number of ops needed to ship new backends; adding backend-agnostic implementations of all secondary ops 3.
SOFTWARE
-
LangSmith Custom Apps builds custom review UIs for agent traces
By
–
every agent team ends up wanting its own review ui over their traces – LangSmith Custom Apps lets you build it on your own data and ship it right into the workspace
-
LangSmith Custom Apps for building interfaces with LLM data
By
–
💻 Build the interface you want w/ your LangSmith data
✅ Publish it into your workspace
🙅 Skip hosting, auth, + permissions work Get started with Custom Apps today. -
Rollouts cloud agent auto-fixes regressions by opening issues
By
–
When Rollouts catches a regression, it finds the offending PR and opens an issue. One click starts a cloud agent to fix it. Rollouts usage credits are included through Oct 3.
-
Enterprise AI Requires Significant Hard Engineering Work
By
–
It has been apparent for a while now that there is a lot of extra engineering work of the plain-old-fashioned hard engineering type, to make AI stuff work correctly in an enterprise setting. I had an interesting chat with Progress Software CEO Yogesh Gupta this week about how the
-
Enterprise AI Requires Extra Engineering Work, Say Dynatrace and Arize
By
–
I have for a while now thought that enterprise artificial intelligence is going to require a lot of extra engineering work of the plain old variety. Some of it can be automated, however, claim Rick McConnell of Dynatrace and Jason Lopatecki of Arize, which just joined Dynatrace.
-
On-device decision models run locally with llama.cpp
By
–
Decision models now run on device in llama.cpp. Free, fast, private! llama serve -hf ggml-org/Kev-4B-GGUF
-
LangSmith Engine v2 validates agent fixes before shipping
By
–
Validate agent fixes before shipping them with LangSmith Engine v2. Engine now tests and validates fixes automatically: ✅ Replicates the issue with the same deployment environment and same inputs
✅ Builds a fix, tests it, and iterates on improving it until it has a -
Python Automation Cookbook with MCP and AI-powered workflows
By
–
Python Automation Cookbook — 100+ new and updated recipes for scalable workflows, MCP integrations, and AI-powered automation: https://
amzn.to/4zc2WOL via @PacktDataML @PacktPublishing -
MCP OAuth token flow between host, server, and embedded UI
By
–
Good question. The host coordinates OAuth, usually opening the provider’s login page outside the iframe. It then sends the access token only to the MCP server. The embedded UI calls authenticated tools through the host, so the token itself is not exposed to the UI.