56 loops of Claude Code just killed every hand-crafted AI attack method. Anthropic just open-sourced a repo called Claudini. It uses Claude Code in an autoresearch loop to automatically discover new adversarial attacks against LLMs. After 56 iterations, it found an
AGENTS
-
SaaS Evolution: Linear Pioneers Agent-Native Software Era
By
–
SaaS isn’t dead, it just needs to become agent-native. Linear (@linear) is a great example of how:
— Dan Shipper 📧 (@danshipper) 1 avril 2026
They pivoted the product to be used by both humans and agents, and that has made them one of the premier software tools in the agent-native era.
I had Linear’s cofounder and CEO… pic.twitter.com/wbgljfcqzXSaaS isn’t dead, it just needs to become agent-native. Linear (
@linear
) is a great example of how: They pivoted the product to be used by both humans and agents, and that has made them one of the premier software tools in the agent-native era. I had Linear’s cofounder and CEO -
Self-Organizing LLM Agents Outperform Predefined Role Hierarchies
By
–
NEW papers on self-organizing LLM Agents. Assign an agent a role, and it'll follow instructions. Let agents figure out roles themselves, and they'll outperform your design. New research tested this across 25,000 tasks with up to 256 agents. The work shows that self-organizing LLM agents spontaneously develop specialized roles without any predefined hierarchy. A sequential coordination protocol outperformed centralized approaches by 14%, agents generated over 5,000 unique roles organically, and open-source models reached 95% of closed-source quality at significantly lower cost. Most multi-agent frameworks today start by defining roles: planner, coder, reviewer, critic. This paper provides large-scale evidence that the opposite approach works better. Give agents a mission, a protocol, and a capable model. The agents will figure out the rest. Paper: arxiv.org/abs/2603.28990 Learn to build effective AI agents in our academy: academy.dair.ai/
-
MemFactory: Unified Framework for Trainable Agent Memory Systems
By
–
// Unified Inference and Training Framework for Agent Memory // Most memory-augmented agents are built with duct tape—one system for storage, another for retrieval, a third for training. New research introduces a unified framework that treats agent memory as a first-class, trainable component. MemFactory provides modular, plug-and-play memory components with native GRPO integration for fine-tuning memory management policies through RL. It supports Memory-R1, RMM, and MemAgent paradigms in one framework, with up to 14.8% relative gains over baselines. Why does it matter? As agents move from single-turn tools to persistent assistants, memory becomes the bottleneck. MemFactory gives researchers standardized infrastructure to build, train, and evaluate memory-driven agents without reinventing plumbing for every new approach. Paper: arxiv.org/abs/2603.29493 Learn to build effective AI agents in our academy: academy.dair.ai/
-
Code-as-Policy Framework for Robot Perception and Control
By
–
Very excited about the prospect of Code-as-Policy (CaP) for Robotics!
— Ken Goldberg (@Ken_Goldberg) 1 avril 2026
Esp with recent rapid advances in agentic coding. CaP has potential to quickly combine VLA models with GOFE primitives into interpretable code, observe experiments, and iterate. Initial results are promising: https://t.co/Z60Q1TwbNkVery excited about the prospect of Code-as-Policy (CaP) for Robotics! Esp with recent rapid advances in agentic coding. CaP has potential to quickly combine VLA models with GOFE primitives into interpretable code, observe experiments, and iterate. Initial results are promising: Max Fu (@letian_fu) Robotics: coding agents’ next frontier. So how good are they? We introduce CaP-X: an open-source framework and benchmark for coding agents, where they write code for robot perception and control, execute it on sim and real robots, observe the outcomes, and iteratively improve code reliability. From @NVIDIA @Berkeley_AI @CMU_Robotics @StanfordAILab capgym.github.io 🧵 — https://nitter.net/letian_fu/status/2039342130565357956#m
→ View original post on X — @ken_goldberg, 2026-04-01 14:25 UTC
-
Every Runs on Notion Agents for Workflow Automation
By
–
we run all of @every on @NotionHQ one of the big reasons we love notion is notion agents. they run 24/7 in the background to help us: – prioritize work
– plan out our strategy
– organize knowledge
– make sure everyone is on the same page on Friday we're doing a Custom -
CaP-X: Open-Source Framework for Coding Agents in Robotics
By
–
Robotics: coding agents’ next frontier.
— Max Fu (@letian_fu) 1 avril 2026
So how good are they?
We introduce CaP-X: an open-source framework and benchmark for coding agents, where they write code for robot perception and control, execute it on sim and real robots, observe the outcomes, and iteratively improve… pic.twitter.com/P547voL0NvRobotics: coding agents’ next frontier. So how good are they? We introduce CaP-X: an open-source framework and benchmark for coding agents, where they write code for robot perception and control, execute it on sim and real robots, observe the outcomes, and iteratively improve code reliability. From @NVIDIA @Berkeley_AI @CMU_Robotics @StanfordAILab capgym.github.io 🧵
→ View original post on X — @berkeley_ai, 2026-04-01 14:00 UTC
-
SIGReg and LeWM Experiments: Challenges with Value Function Generalization
By
–
Adding a little bit of SIGReg to prefinal activations did coerce them into independent Gaussian, but it hurt generalization on value functions. Training a full LeWM ahead of time also resulted in worse value function estimation. I’m not giving up yet, but my first few attempts
-
LaPha: AI Agents Think Exponentially Better in Poincaré Space
By
–
How can we give AI agents exponentially more room to think and solve complex problems? Researchers from Shanghai Academy of AI for Science, CMU, and others unveil LaPha. This new method trains AlphaZero-like LLM agents in a unique "Poincaré latent space." It leverages negative
-
5 Production Scaling Challenges for Agentic AI in 2026
By
–
5 Production Scaling Challenges for Agentic AI in 2026 machinelearningmastery.com/5…
→ View original post on X — @craigbrownphd, 2026-04-01 12:44 UTC