I’ve launched Headlamp Theme Builder.
It’s a small tool for creating custom themes for Headlamp, with a live preview so you can see how colours, surfaces, navigation, buttons, terminal colours, and logos will look before using them.
You can use it in two ways:
Website: build and download a theme plugin
Headlamp plugin: build, preview, and apply a theme directly inside Headlamp
It also includes a theme library with ready-made palettes like Pixelrobots, GitHub, Dracula, Nord, Solarized, Monokai, and cloud-provider-inspired themes.
Try it here:
https://t.co/rT8lIS6YZM
GitHub:
https://t.co/JwiFtHU7N1
1000 agents, 50 active at any moment.
You pay for ~50 worker pods, not 1000 Agents.
Idle agents cost a valkey row (key-value data store and cache) + GCS bytes for the snapshot.
Actors (Agents) are only active when they're being used. Otherwise, they're suspended.
#agenticai #ai
Trying out the new @GitHub Copilot App with @OpenAI codex! Lets see how many #AIAgents I can get cooking at the same time and how many tokens I can burn! 🔥
🚀 Transform Your Production Line with Agentic AI
https://t.co/zdA85EZkq6
Unlock the future of manufacturing with Agentic AI in Manufacturing, a cutting-edge course designed to help you streamline operations, cut costs, and boost efficiency using the power of agentic AI systems.
When I talk to folks who are building internal productivity apps, they often assume those apps don't require the same rigor as production software.
Fam, these apps are connected to your company's systems and data... they ARE production apps!
So, you still need security, and governance, and all of the other things that are necessary but kinda suck the joy out of building 🫠
But what if you could still have fun while also being a responsible adult?
Retool lets you safely color inside the lines. It bakes all of that grownup stuff in while you focus on the fun parts of building.
Check out the Retool platform to learn more about how you can quickly build apps that are secure and enterprise-ready from day one:
🖍️ https://t.co/oJIIh5rk7O
In collaboration with @retool
I spent 6 years working on the ServiceNow platform.
Now I’m diving deep into Microsoft Azure OpenAI Services, Azure Machine Learning, Foundry, Microsoft Copilot Studio, and the Power Platform.
One lesson I’ve learned in tech: don’t tie yourself to just one platform. The industry evolves fast and so should you.
Are your business APIs ready for today's Agent-first ecosystem?
Over the years, companies have invested heavily in building and maintaining APIs. Now, AI Agents are the new consumers - and they speak #MCP, not REST
In this short demo, I show how to:-
1. Import your API into #AzureAPIManagement,
2. Convert it into an MCP Server,
3. Test it in #VSCode with #GitHubCopilot
📣 Claude Fable 5, the first in @AnthropicAI's Mythos model class, is now generally available and rolling out in GitHub Copilot.
It is designed for long-horizon, autonomous coding and knowledge-work tasks. Try it out in @code or the GitHub Copilot app. ⬇️
https://t.co/jJTqh35jjY
Anthropic’s newest model, Claude Fable 5, a Mythos class model for non-security work lands in Microsoft Foundry and GitHub Copilot today.
Read the blog here: https://t.co/Z48DbB7csq
🚀 New repo drop: AI Agents for IT/Ops
Hands‑on samples + patterns to help IT/Ops teams manage AI agents. Not about building agens, but how to manage them!
🔗 https://t.co/2YKTDqhdAT
Practical. Open-source. Built for real IT/Ops scenarios. More to come soon!
#AzOps#ITPro
MICROSOFT OPEN-SOURCED THE COMPLETE MCP PLAYBOOK
Every ai engineer needs to understand model context protocol in 2026
Microsoft built a full open-source course so you don't have to figure it out the hard way
here's what's inside:
▫️ 11 modules from zero to production
▫️ hands-on labs in python, typescript, java, rust, c# and javascript
▫️ covers mcp servers, clients, security, oauth2, azure integration
▫️ a 13-lab capstone with real postgresql + vector search
▫️ works with claude desktop, cursor, vs code and more
▫️ 16k stars. 5.3k forks. built by microsoft.
the curriculum even teaches adversarial multi-agent reasoning two agents debate using shared mcp tools, judged by a third agent.
if you're building with ai agents in 2026, mcp is the layer that connects everything. this is the fastest way in
https://t.co/jZlCZE8VkX