Tired of LLMs forgetting everything? The goldfish problem is real. A new blog post from Emerging Technologies explores architecting memory for AI agents, enabling long-term learning and "enterprise mind" systems. https://t.co/duSLBw6GsV
Zero Trust for AI Agents: Stop letting AI agents inherit full user permissions. Discover how a delegation model with permission intersection enforces Zero Trust on OpenShift. https://t.co/8Y8Jn9NZvj
Benchmarking: Engineering teams need a rigorous, repeatable method for evaluating CPU performance. Enter the vLLM CPU Performance Evaluation framework for benchmarking AI inference, developed by Red Hat Emerging Tech. Learn more https://t.co/xisYT225Ro
Want to experience two days of expert talks, networking, and great food in the sunny surroundings of @FIT_VUT ? ☀️
Hurry and register for #DevConf_CZ 2026 now, spots are filling up fast! 🏃♂️💨
🎟️ https://t.co/etj2HHlF20
Who is really calling your AI agents? 📷 The gap between what an agent claims and what the platform can verify is a huge attack surface. Here's a 3-step, zero-trust approach to securing agent-to-agent communication in a Kubernetes cluster. https://t.co/8GMYhCpvbS
Introducing an agentic OS prototype that combines:
🔄 Fedora-bootc for OS lifecycle and transactional updates
📦 Rootless Podman for workload isolation
🦞 OpenClaw for a robust agent runtime
Check out the demo and start building: https://t.co/exWPeyQYzA.
Stop context bloat! Red Hat's codemode-lite uses sandboxed Python to cut massive scaling bottlenecks by reducing tool schema overhead by 53% on identical tasks in tests, and savings increase with scale. Learn more on Next: https://t.co/zRqaGJflha
Red Hat has collaborated on the Gateway API Inference Extension within the #K8s community! See how this project streamlines the serving process for LLMs with complex computational profiles. #Kubernetes#GatewayAPI#Inference#MachineLearning https://t.co/3JFDEReTf3
PyTorch Community RTP – April Meetup
We’re back with our second meetup on April 28 after a great kickoff!
This time we’re diving into system-level ML — how training, kernels, and deployment come together in real-world systems. RSVP on Meetup: https://t.co/k9pSg0lBLg
Red Hat supports open source projects for Sovereign AI, letting you run large models like DeepSeek w/in your own infrastructure. Tools like Dockling, VLM, & llm-d help you run models and scale across clouds. #SovereignAI#OpenSourceAI https://t.co/oC2GH4noqh
Training models on your own data with open source tech builds trust in the output. Projects like vLLM and llm-d make AI scaling more energy efficient and cheaper by optimizing inference across many systems. #OpenSourceAI#Trust#ScalingAI#VLLM#llm-d https://t.co/YgaVkhJNtu
The more we can reproduce open AI, the more progress made in areas like #healthcare, #finance, & #education. Red Hat enables "sovereign AI" by bringing tested, verified open source projects to enterprises. #OpenSource#RedHat#Transparency https://t.co/zk9XIVBGUX
The cloud-native community offers key lessons for the #AI era. The core challenge: "too many projects, not enough reviewers." Make open source software production faster and easier with intelligent automation. #OpenSource#CloudNative#Automation https://t.co/gM5N2eE4KG
Meet contributors, committers, working group leaders, and architects from across the ecosystem at #OCX2026. Exchange ideas, solve problems, and build relationships that continue long after the event.
True #OpenSourceAI is more than just open code—it requires open architecture, open weights, & open data.
We need vibrant communities & open governance around models,as the AI landscape expands and agents self-learn.
#AI#CommunityDriven#AITransparency https://t.co/XjBXHNWC2y
Quality over quantity! 🚀 AI helps write code fast, but don't overwhelm maintainers with unverified PRs. Focus on meaningful contributions instead of dozens of notifications. Respect community, stay human, stay helpful! 🤝 #OpenSource#AI#Programming https://t.co/bTvOtifaI1
One ticket. Full access to five focused collocated events:. Move between #OCX2026 and five domain-focused communities. Go deep where it matters most to you. https://t.co/tTmQV8k3jQ
Quality over quantity! 🚀 AI helps write code fast, but don't overwhelm maintainers with unverified PRs. Focus on meaningful contributions instead of dozens of notifications. Respect community, stay human, and stay helpful! 🤝 #OpenSource#AI#Programming https://t.co/qNHJxoGa4Y
AI can halve your workload by generating boilerplate code, but accountability remains key. Always test contributions, because you—not the AI—are responsible for the quality and licensing of the code you submit. #AI#SoftwareDevelopment https://t.co/x2N5X4uOET
AI can halve your workload by generating boilerplate code, but accountability remains key. Test contributions and ensure you can explain every line, because you—not the AI—are responsible for the quality and licensing of the code. #AI#SoftwareDevelopment https://t.co/bDo02ZApRD