@LangChainAI I presented a talk around Agentic Auth at @KubeCon_ around building secure AI Agents using @OpenFGA

@LangChainAI This is a great initiative by Langchain

Market dips can mess with your college fund. 💡Here’s how to safeguard your 529.

@LangChainAI Agent Auth, now this is something interesting, good job team @LangChainAI 👍

@LangChainAI nice, more concerns for secops

@LangChainAI Securing agents is crucial as they evolve. How do you envision balancing their growing capabilities with robust security measures? Interested in any examples of innovations tackling these challenges!

@LangChainAI I was thinking of adding this feature to my SaaS but thanks to you, I’ll just use the middleware

@LangChainAI Something interesting !!

@LangChainAI yeah reliable agents save headaches for small crews

@LangChainAI Securing access is often overlooked, and this guide makes OAuth & OIDC practical for real agent architectures.

@LangChainAI Useful great

@LangChainAI If token-based least-privilege isn't your baseline, you're building a sandcastle. Any agent stack relying on human-scale MFA for auth is, by definition, pre-production. Your agents need a key, not a text message from their human parent.

@LangChainAI Agent security is the next big problem. When your agent can call APIs, update databases, and send messages autonomously, auth/authz becomes critical. Traditional API security doesn't cut it. Need agent-specific patterns for action verification and scope control.

@LangChainAI agents are the new actors, security is the new challenge

Get Rich Scheme For Techies - practice training small LLMs - produce a small LLM (3B model) that is SOTA - use common crawl data for pre-training - perform some basic fine-tuning with open-source, good-quality datasets - make a post on X and show off your benchmarks - create a

Introducing Mistral AI Studio, the production AI platform. Mistral AI Studio enables builders to move from AI experimentation to production with a robust runtime for agents and deep observability across the AI lifecycle. More on our blog: https://t.co/acTq0MZgld

This is how I structure our multi-agent AI system: • 10 agents (triage + 9 specialists) • 43 tools (grouped by domain) • 12 artifacts (visual canvases) Each agent gets only the tools it needs. Clean and maintainable.

Qwen is now making 51% in a week. 20x leverage, all in BTC - forceful, blunt, but effective 😅 DeepSeek built more nuanced portfolio (mostly long) and has been consistent on 20-30% returns. All other models are losing money, with ChatGPT lost 72%. Fascinating experiment.