5/5 The takeaway: If your agent relies on an LLM judge for selection accuracy, measuring code quality isn’t enough; you need a measure of the model's inductive bias toward the "fingerprint" of a gold solution. This was our blueprint.
AGENTS
-
LLM Judge Rejects Functional Fix for Code Aesthetics
By
–
3/5 An example: In instance psf__requests-1724, the gold fix is 2 lines. Our agent’s functional fix was 8 lines. The LLM judge rejected the correct 8-liner as "messy" and "redundant," choosing a clean but **non-functional** fix instead. See full patch in the blog:
-
Claude Opus 4.5 Judges Gold Patches Without Memorization
By
–
1/5 When we saw our Reducer (=LLM judge component in Maestro, our agentic framework, that selects the best output from parallel agent runs) consistently picking gold patches, we were sure Claude Opus 4.5 (knowledge cutoff Aug '25) had simply memorized the answers. But then we
-
LLM judges reject functional agent fixes as messy code
By
–
3/5 An example: In instance psf__requests-1724, the gold fix is 2 lines. Our agent’s functional fix was 8 lines. The LLM judge rejected the correct 8-liner as "messy" and "redundant," choosing a clean but **non-functional** fix instead. See full patch in the blog:
-
Using AI Agents for Automated Email Management
By
–
The worst part of building isn’t shipping.
— God of Prompt (@godofprompt) 15 avril 2026
It’s the pile of emails, follow-ups, and random ops work waiting for you right after.
Been using Wingman in beta and waking up to a cleaned up inbox + drafted replies is hard to give up now. https://t.co/CThyrFrWywThe worst part of building isn’t shipping. It’s the pile of emails, follow-ups, and random ops work waiting for you right after. Been using Wingman in beta and waking up to a cleaned up inbox + drafted replies is hard to give up now.
-
GetDiana: Enterprise-Grade AI Counterpart for Team Collaboration
By
–
🚨 What happens when you give your employees an employee?
— Charly Wargnier (@DataChaz) 15 avril 2026
Say hello to @GetDiana: the new, enterprise-grade OpenClaw.
She just launched to give every single person on your team their own dedicated AI counterpart.
She lives right inside your Slack DMs and handles the heavy… pic.twitter.com/o715YBFMlWWhat happens when you give your employees an employee? Say hello to @GetDiana
: the new, enterprise-grade OpenClaw. She just launched to give every single person on your team their own dedicated AI counterpart. She lives right inside your Slack DMs and handles the heavy -
LangSmith Fleet Enables Team Agents With Admin Controls
By
–
Fleet makes it easy for anyone on your team to build and use agents. These controls give admins visibility into what agents do and what they can access. Try LangSmith Fleet:
-
LangSmith Fleet: Tool Access Controls and Usage Tracking
By
–
New in LangSmith Fleet: Tool access controls and usage tracking.
— LangChain (@LangChain) 15 avril 2026
📊 Track cost and usage by user, agent, and tool from a single dashboard
💳 Set spend limits per user or team to prevent surprises
✅ Control which tools each user can access with role-based and attribute-based… pic.twitter.com/BBDPaDUxmJNew in LangSmith Fleet: Tool access controls and usage tracking. Track cost and usage by user, agent, and tool from a single dashboard Set spend limits per user or team to prevent surprises Control which tools each user can access with role-based and attribute-based
-
Trusted Tester Shares AI Platform Experience and Publishing Window Challenge
By
–
okay 🙂 im in trusted testers for whatever thats worth. wanted to make it title story for ainews, but window is gone as a consequence of the gradual rollout. just showing my side of things.
-
NVIDIA Nemotron 3 Super: 120B Open Model for Agentic Reasoning
By
–
Banger paper from NVIDIA. Agentic reasoning needs models that are not just capable, but efficient at long-context inference. The agent model layer is moving toward open, long-context, high-throughput architectures. This paper introduces Nemotron 3 Super, an open 120B parameter