Pinned Tweet
The CFP for AGNTCon + MCPCon is open! 🎉 Come share what you're learning building with agents.
Submit by January 18. events.linuxfoundation.org/a…
Angie Jones retweeted
new Web Dev Challenge at 3pm Pacific today 👀👀👀👀
controlling web apps with hardware sensors — the devs went off on this one
youtube.com/watch?v=WmQr_Yvv…
Angie Jones retweeted
MCP is becoming event-driven.
Something interesting happened in the MCP ecosystem today.
MCP Events is now in beta.
Until now, most MCP interactions have followed a request-response pattern.
An agent needs something, it calls a tool, the server responds, and the agent continues.
But think about how most real-world applications work.
Things happen without us asking for them.
An email arrives. A payment succeeds. Someone opens a GitHub issue. A production server goes down.
Wouldn't it be better if an agent could subscribe to these events and get notified when they happen?
That's exactly what MCP Events is trying to make possible.
Instead of constantly asking a server whether something has changed, clients can subscribe to event types and receive updates through different delivery mechanisms:
- Polling: Check for new events.
- Push: Receive events as they become available.
- Webhooks: Get events delivered through a webhook endpoint.
And this opens up some interesting possibilities.
Imagine an agent that monitors your inbox and prepares replies when important emails arrive.
An agent that watches your GitHub repositories and investigates new issues.
Or an agent that monitors production incidents and starts investigating as soon as something goes wrong.
These are just a few examples.
I think this is an important step for MCP.
We've been making it easier for AI agents to interact with different applications. But interaction isn't just about asking questions and executing actions. It's also about knowing when something happens.
That's a big part of building agents that can operate continuously rather than waiting for someone to prompt them.
One thing worth remembering: receiving an event doesn't automatically authorize an agent to take action. Event payloads are also untrusted input, so security and permission boundaries still matter.
MCP Events is still in beta, and I'm looking forward to seeing how clients and servers start adopting it.
@AgenticAIFdn
@aiDotEngineer
Angie Jones retweeted
Why do voice agents play fake keyboard sounds while processing a request?
Auditory cues help manage perceived latency, making the wait feel more natural even when the agent is still working.
Erica Hughberg and @Dpbrinkm break it down.
Angie Jones retweeted
mcp events are here!!!! 🚀 the beta extension went final TODAY
think of it as webhooks for mcp 📲 your agents an be proactive now with mcp!
all a user has to do is prompt like: “alert me when we have a p0 bug"
chatgpt supports webhook mode today- in work chats on web and desktop
how it works:
1/ the host (like chatgpt) asks your server which events it offers
2/ a user prompts like “tell me when a p0 happens”
3/ the host subscribes and gives your server a callback url
4/ (something happens: an incident opens upstream)
5/ your server sends a POST request w/ a signed event to that url
6/ the host starts an agent turn with the user’s instructions
7/ the agent calls your tools and posts a summary in chat
if you’re building for it rn- id treat event payloads as untrusted ofc and do check where chatgpt expects the capability (capabilities.events in openai docs / capabilities.extensions in the spec)
more links in the replies
The MCP Events extension is now finalised in beta. This means it is ready for clients and servers to adopt and provide feedback, so that we can fix any spec issues and get it stabilised.
modelcontextprotocol.io/exte…
Thanks to all the working group members, maintainers, and early implementers for getting this over the line. Very excited to see what people build with it :-)
(and here's a cheesy video that on one asked for)
Angie Jones retweeted
This is huge! MCP Events are here!
This makes your agents proactive and react and respond to the outside world...
with no complicated setups, or crazy webhook configurations, all it takes is a single prompt in natural language,
I am already testing it and working on support in Postman for all the 3 delivery methods including webhooks.
The MCP Events extension is now finalised in beta. This means it is ready for clients and servers to adopt and provide feedback, so that we can fix any spec issues and get it stabilised.
modelcontextprotocol.io/exte…
Thanks to all the working group members, maintainers, and early implementers for getting this over the line. Very excited to see what people build with it :-)
(and here's a cheesy video that on one asked for)
Angie Jones retweeted
The MCP Events extension is now finalised in beta. This means it is ready for clients and servers to adopt and provide feedback, so that we can fix any spec issues and get it stabilised.
modelcontextprotocol.io/exte…
Thanks to all the working group members, maintainers, and early implementers for getting this over the line. Very excited to see what people build with it :-)
(and here's a cheesy video that on one asked for)
I used @AgenticAIFdn projects agentgateway, OpenSandbox, and @KeycardAI's RFC 8693 and Cedar to build a secure software factory that uses open source and open standards.
Angie Jones retweeted
Open intelligence needs all the help it can get.
Spiral has $$ to fund non-commercial open source AI projects that make intelligence more useful, open, and secure. We’re looking globally.
Apply for a grant: spiral.xyz/grants/
Read more ↓
Angie Jones retweeted
“An unmissable event for anyone serious about production AI agents.”
Join us at AGNTCon + MCPCon North America 2027, April 28–29 in San Francisco!
🎟️ Early bird tickets are just $399 for a limited time.
Get your ticket: bit.ly/4vkntyD

ALT The image is a promotional graphic for AGNTCon + MCPCon North America, scheduled for April 28-29, 2027, in San Francisco, California. It features a quote: “AGNTCon + MCPCon is officially where the future of the agentic stack is being built. An unmissable event for anyone serious about production AI agents.” The background is blue with geometric patterns and includes a robot illustration. The tagline reads, "Where the agentic stack is being built.
Angie Jones retweeted
Jesse Vincent ran 100 billion tokens in a month. He also saw AI code reviews stretch from 3 rounds to 80.
He discusses agent harnesses, review loops, and knowing when to stop fixing bugs.
🗣️AGNTCon is baaaaack!
Registration is OPEN for AGNTCon + MCPCon North America 2027!
Early bird tickets are just $399 for a limited time! Don’t miss your chance to save.
Join the builders shaping what’s next in agentic AI, April 28–29 in San Francisco.
🎟️ Grab your ticket: bit.ly/4vkntyD
Angie Jones retweeted
Registration is OPEN for AGNTCon + MCPCon North America 2027!
Early bird tickets are just $399 for a limited time! Don’t miss your chance to save.
Join the builders shaping what’s next in agentic AI, April 28–29 in San Francisco.
🎟️ Grab your ticket: bit.ly/4vkntyD
Angie Jones retweeted
Excited to see MCP Events to land in the spec: unlocks a whole new set of push-based workflows between services and agents. 🎉
Bonus, we've implemented and support current profile on Shop MCP (shop.dev): the MCP client can register for push based order updates on behalf of authenticated user! Take it out for a spin.
Nah the vast majority of the time, it's slop. And I'm so sick of reading Claude generated plans
A new faux pas I’m seeing: dismissing a teammate’s well-structured, probably decent agentic output as “AI slop” because you don’t want to read it.
If a teammate (or agent) has skillfully prompted and carefully edited something to solve a problem or communicate with the business, it isn’t automatic slop just because AI generated it. There’s often good stuff in there... if it’s long or you're multi-tasking, feed it to your agent.
Don’t overuse the term "AI slop" for any AI output. Save it for the actual sloppity slop, ya know?
Angie Jones retweeted
BREAKING: Google says “the era of autonomous attacks has arrived.”
Hackers are now using AI to scan software updates, figure out which security flaw each patch fixes and build an attack before companies can install the fix, all without a human involved, according to a new report from the Google Threat Intelligence Group.
“Defending against machine-speed threats at human speed isn’t feasible,” Google said.
Angie Jones retweeted
I am so excited to lead the Developer keynote at AWS reInvent this year!
Working on the talk track atm, and we've got such a great lineup of speakers, deep dives, and practical advice
Sign up for the livestream here:
aws.amazon.com/events/reinve…
Monday, Nov 30, 7:30 pm - 8:45 pm
great talk from @jpschroeder making the case for why companies should build small domain-specific agents
youtu.be/wIPr9GmLRaQ
Angie Jones retweeted
What does software engineering look like when generating the code becomes cheap and verifying it is the hard part? A rare podcast focused on everything around building software, with @satyanadella. Timestamps:
00:00 Intro
02:12 Microsoft’s open-source shift
04:15 Building VS Code in the open
05:55 Copilot, Cursor, and Claude Code
09:52 Decision engines
12:32 Verifying AI-generated code
15:39 The software factory
20:45 Bringing AI to existing products
24:06 Why adopting AI is different to adopting React
26:29 The changing role of software engineers
29:25 Microsoft’s voluntary exit offers
34:05 AI anxiety and changing roles
36:35 Junior hiring
42:36 Open models
45:39 Owning your AI infrastructure
51:21 Building trust in AI agents
53:48 Local models and hybrid intelligence
56:39 Microsoft’s model strategy
59:26 Advice for engineers
1:01:21 Closing takeaways
Brought to you by:
• @turbopuffer – a vector and full-text search engine built on object storage. It’s fast, cheap, and extremely scalable. turbopuffer.com/pragmatic (I built a greatly improved podcast/newseltter search with it, check it out: pragmaticengineer.com/search)
• @linear – the product development system for teams and agents. linear.app/pragmatic
• @WorkOS – everything you need to make your app enterprise-ready. workos.com/
3 interesting takeaways from Satya:
1. A new form of software engineering: teams building software factories to improve their work.
Satya cites the Azure networking team which manages 500+ fiber network operators and recently built an agentic system to better manage emails and processes. Now, they operate much more efficiently as a result. Satya sees this as the reality of “AI transformation”: brownfield work on building software factories to improve how teams operate, all while continuing the day-to-day work.
2. “Own the intelligence”: avoid lock-in & treat agents as an insider risk.
Satya cautions companies to avoid lock-in with any single model provider by ensuring that the harness, memory, and model are not coupled, so that they can be changed with ease. He also thinks more companies should consider using open models as a way to protect their core intellectual property (IP) from leaking out, and build private evals to decide what models to use and upgrade to. He also says that models are intelligent but unreliable, so it’s best to treat them as “insider risks” and build a separate system to verify that agentic systems work as expected.
3. What’s with those voluntary exit offers?
I asked Satya directly why Microsoft recently made voluntary exit offers to most of its software engineers for the first time in the company’s history. I also wanted to know if it could mean layoffs in the future. Unsurprisingly, his reply was diplomatic, but in the podcast I rehashed what I heard in a more direct way.





























