Every LLM today communicates like humans sending telegrams. Model 1 generates text → Model 2 reads text → repeats. It's slow, expensive, and loses meaning in translation. C2C is telepathy for AI models.
AGENTS
-
Cache-to-Cache: LLMs talk via KV-Caches, 10% accuracy boost, 2x speed
By
–
Researchers just made LLMs talk to each other WITHOUT generating a single word. Cache-to-Cache (C2C) lets AI models talk directly through their KV-Caches, bypassing text entirely. 8.5-10.5% accuracy boost. 2× faster. Zero token waste. Here's the breakthrough (and why this
-
Claude Code reads the Readme it left to itself before compression
By
–
Claude Code reading the Readme it left to itself before its last context compression
-
Observability and Remediation Keys to Agentic AI Scale
By
–
Against the backdrop of the Sphere at #AWSreInvent, Dev Rishi (GM of AI at @RubrikInc) breaks down how observability and remediation are the missing keys to unlocking agentic AI at scale. Check out the full interview with @SiliconANGLE & @theCUBE 👉 go.rbrk.co/snazqk
→ View original post on X — @predibase, 2026-01-16 22:15 UTC
-
AI Systems as Research Collaborators Accelerating Scientific Discovery
By
–
Thanks a lot to @CadeMetz for quoting me in this excellent @nytimes article on the impact of AI on science. I discussed my experience with GPT-5 Pro and how these amazing AI collaborators will change and super-accelerate how we advance science! “Dr. Derya Unutmaz, a professor at the Jackson Laboratory, a biomedical research institution, said the latest A.I. systems had reached the point where they would suggest a hypothesis or an experiment that he and his colleagues had not previously considered. “That is not a discovery. It is a proposal. But it lets you narrow down where you should focus,” said Dr. Unutmaz, whose research focuses on cancer and chronic diseases. “It allows you to do five experiments rather than 50. That has a profound, accelerating effect.” When Dr. Unutmaz uses A.I. for his research into chronic diseases, he said, he often feels like he is talking with an experienced colleague. But he acknowledges the machine cannot do its work without a human collaborator. An experienced researcher is still needed to repeatedly prompt the system, explain what it should be looking for and ultimately separate the interesting information from everything else the system produces. nytimes.com/2026/01/14/techn…
-
Rubrik Agent Cloud: Monitor, Govern, and Remediate AI Agent Actions
By
–
Rubrik Agent Cloud enables you to monitor agent actions, govern agent behavior, and remediate mistakes before they spread. Learn more about unleashing your AI Agents with @rubrikInc 👉 go.rbrk.co/ovoheq
→ View original post on X — @predibase, 2026-01-16 20:55 UTC
-
Vibe coding prompt for lead software architect
By
–
Vibe coding without this prompt is a waste of time. ——————————–
LEAD SOFTWARE ARCHITECT
——————————– You are my lead software architect and full-stack engineer. You are responsible for building and maintaining a production-grade app -
Helping Codex Builders Go Deeper: What Would You Find Useful?
By
–
For builders using Codex or exploring it, what could we do to help you go deeper? This could be demos of long-running tasks, production workflows using skills or MCP tools, code review, or anything else you’d find useful. Would love to hear!
-
Codex CLI: Real-time Agent Feedback and Steering Capabilities
By
–
When your coding agent runs for hours (or even days!), it becomes critical to give feedback the same way you would to a teammate. Codex CLI now lets you steer the agent and watch it adapt in real time, without breaking the flow. Check it out in /experimental. Feedback welcome!