Register for Open Source SecurityCon North America 2026

What’s in the SOSS? Podcast #74 – S3E26 Building the Agentic Future with Angie Jones

By September 29, 2026Podcast

Summary

In this episode of What’s in the SOSS?, host Sally Cooper interviews Angie Jones, VP of Developer Experience at the Agentic AI Foundation (AAIF). Angie shares her career journey from software engineering and holding 27 patents at IBM to leading open source protocol and developer relations initiatives. She details the mission of the Agentic AI Foundation as a neutral home for governing open source standards and technologies such as the Model Context Protocol (MCP), Goose, Agents.md, and Google’s Agent-to-Agent (A2A) protocol. The conversation explores how cross-industry collaboration addresses the reliability and security of AI agents in production, key considerations for the shift toward agentic engineering, and what attendees can expect at the upcoming AgentCon and MCPCon North America.

Conversation Highlights

00:00 – Introduction
00:28 – Welcome Angie Jones
01:47 – Angie’s Career Journey: From IBM Patents to Open Source
06:56 – Mission of the Agentic AI Foundation & Governing Open Standards
11:47 – Preparing for the Era of Agentic Engineering
13:20 – Preview of AgentCon and MCPCon North America
17:32 – Rapid Fire Questions
18:36 – Wrap-up & OpenSSF Resources

Transcript

Intro & Music (00:00)
It becomes a point in time in an open source project or standard’s life where it’s like this is bigger than the company that created it. And so that’s what Agentic AI Foundation is, That neutral home where these critical technologies can live.

Sally Cooper (00:28)
Hello, hello, and welcome to What’s in the SOSS? The OpenSSF podcast, where we get to talk to amazing people and talk about the ingredients that work together to secure open source software. I’m your host, Sally Cooper. And whether you’re building software or just curious about the future of agentic AI, we have a fantastic guest for you. Joining me today is Angie Jones.

She has a very impressive resume, VP of Developer Experience at the Agentic AI Foundation. Angie, I know you guide how agentic systems are designed, built, and adopted across a developer community. You’re an award-winning teacher, an international keynote speaker, and I have to say I’m a huge fan girl. I saw your SEO changing in this new world of AI at the LF All Hands, and it has really helped me in marketing and communication. So I’m just so excited to talk to you today. Thank you for being here.

Angie Jones (01:34)
Yeah, thanks so much for having me, Sally.

Sally Cooper (01:37)
Aw, thanks, Angie. So let’s get started. Tell me a little bit about your journey and your role at the Agentic AI Foundation.

Angie Jones (01:47)
Yeah, you know, it’s really funny. Sometimes I sit and think, like, how did I get here? How did I end up here? You know, it’s like I’ve become a protocol girly and when I reflect over my career, I can see how that happened. And I think that, you know, all of my various roles have kind of led up to this moment. So I have a traditional tech background, you know.

Study computer science, you know, bachelor’s and master’s degree in that. I worked at IBM as a Software Engineer is where I started my journey. And at some point I got really into automation. So software automation, test automation, just automating things. And this was before it became cool with AI, you know, this was well long ago. And you know, throughout my career I’ve

For just automating things, building quality software. I felt so passionate about that that I just started blogging about it. And, you know, I would volunteer to go and speak at tech conferences, and then I found that hey, I really love this sort of thing, you know. I even worked as an adjunct professor, teaching computer science programming classes.

And so all of that kind of teaching and sharing prepared me for a role in Developer Relations. and so I kinda switch. I don’t say I switched from engineering to DevRel, because DevRel is very much an engineering role as well. It’s just adding the teaching component to it. and so all of that then led me to open source.

So like when I was at IBM,we were all about just being innovative. Like it’s just in the company DNA. so there I just always stayed on the cutting edge of technology, ended up with 27 patents at IBM. And so this was my first job as well. And so this has definitely laid the foundation for what I like to do and like how I like to basically shape the future of technology. And so that’s followed me along. I worked at an AI startup and this was pre Gen AI, right? So before Chat GPT and all of that, I was already working at this AI startup and, you know, kind of exploring that and machine learning. Then eventually went to Block where I actually worked at Twitter before the AI startup. And so Jack Dorsey was the CEO of Twitter and also the CEO of Block. And they were starting this new open source protocol for cross-border money movement. And so Jack tapped me to come over to Block to lead that initiative. And so that was like my first plunge into like protocols and stuff like that. I was already really involved in open source and stuff.

At Block, I then started leading Developer Relations for the entire company. So not just that protocol, but all of our products, including our open source ones. And Goose was born. So this was then when the models had two calling features added to them. And so I started leading DevRel for Goose, which is an open source AI agent, one of the very first ones in the world actually and so i found myself wearing all of these hats the the teaching hat and everything where i needed to essentially teach the Developer community what is an AI Agent right that’s how early we were with this and so Block, Anthropic, and OpenAI then founded the Agentic AI Foundation so i was a part of that founding and felt so passionate about the mission. I then came over to help lead the Agentic AI Foundation. So here I am doing all the things cutting edge technology, open source, protocols, and AI. So a perfect mix.

Sally Cooper (06:21)
It is a perfect mix. Thank you for your origin story. I’m even more blown away. Wait, first of all, 27 patents.

Angie Jones (06:29)
Yes, yes.

Sally Cooper (06:31)
That’s incredible. What a great story, Angie. I love getting to know the people behind the work. At OpenSSF, so much of what we do is bringing people together. So that’s why I felt like this was a natural fit. Just thinking about, you know, our mission is to secure open source software, right? Tell me a little bit about AAIF’s mission. And then let’s think about how we can naturally connect.

Angie Jones (06:56)
Yeah. So AAIF was formed out of necessity. So there’s like all of this innovation that’s happening very quickly in this space, right? You have new labs that have sprung up and are leading the path, like Anthropic and OpenAI, right? and things like MCP, for example, is a Model Context Protocol.

This was created to help agents connect to our systems that we use our tools so that they can get the data that they need and the context they did need to be able to perform our tasks, right? And while this was born in Anthropic, the adoption is across the board. You know, at AAIF, I have the opportunity to visit various countries and, you know, talk to different companies and across the board, you know folks are adopting this protocol. And so Anthropics saw that need themselves, and that hey, this can’t live with us, right? We can’t be the sole owners and operators of this shared protocol. Everyone is using it, multiple companies are contributing to it. and so it it becomes, you know, a point in time in a open source project or or standards life where it’s like this is bigger than the company that created it. and so that’s what a Agentic AI Foundation is, that neutral home where these critical technologies can live, right? And so this is where they’re governed and companies feel really good and reassured by this because you don’t really want to build your business on top of something that you know is owned by one company and it’s only following their roadmap, and you know, maybe they can change it, they can pull it, they can delete it or whatever, right? And so this is something that the community kind of demanded, if you will. and Anthropic was totally on board with that.

In addition to MCP, we also have other critical standards, Agents MD, which you know everyone is using in their software development projects. We have Goose, which is the open source agent that I mentioned. We have A2A, which is Google’s agent to agent protocol that just became a member. And we also have a couple of agentic gateways as well. And so we are building the agentic stack. We like to say this is where the agentic stack is being built because as all of these companies and communities, some of them are not even companies, you know are innovating and are solving the new challenges that we’re finding with agents basically doing work for us, these things need a neutral home. And so that’s what the Agentic AI Foundation is, and it gives this place for companies to collaborate and figure this stuff out together.

So we have multiple working groups that are comprised of competing companies that sit together, you know, every week, every two weeks, and they figure out together how these things should work. For example, one of my favorite working groups is the agentic commerce one, right? Because people love the demo of, I’m gonna have my agent buy something for me from Amazon or whatever. But there’s a lot that goes into enabling something like that, you know, the shopping experience, the authorization, the identity of the agent and the human, right? What are they allowed to do? How are they actually gonna buy her? You’re giving them your credit card, you shouldn’t do that sort of thing, right? So, all of these various considerations are things that you don’t want one company making these decisions, right? And so you have, you know, the PayPal’s, the Shopify’s, the Visas of the world together at one table defining, I like to say, the the game rules together and then they can go and compete on top of these standards.

Sally Cooper (11:20)
Well said, Angie. I love how you laid that out. And thinking about the community and the working groups and just so much that they can accomplish together. I had one other question for you, just about developers and organizations as they consider agents moving into production. And yeah, tell me like what should what are the main things that they need to know?

Angie Jones (11:47)
Yeah, you know, I personally think that right now it’s only the beginning for what developers will do with agents. Right now, you know, most developers are looking at Claude codes, the codexes, you know, the coding agents. And I think that’s just the tip of the iceberg. What I envision is going to happen is software engineering is gonna become more of agentic engineering. You’ll need to develop agentic systems, right? And there’s so much that goes into doing that, you know. Some of the things we already have the skills to do, like systems thinking and stuff like that. But there’s a lot of new things and components as well that we need to consider on how do you make a system reliable?

For example, right? So you send an agent, which is a non-deterministic system, to go complete some task. How do you know that it’s going to do that right? And this is what engineers need to focus on is how do I build a system, a reliable and secure system around a model, so that one, it does what it needs to do, and two, if it does go off the rail, I have guardrails to kind of catch that.

And three, that I’m able to observe all that’s happened and be able to improve the system based on those observations.

Sally Cooper (13:20)
I love that. And I’m thinking about our communities. It’s just great to have you here to talk about this on the podcast. But there’s also a great opportunity coming up to have these conversations in person. Our OpenSSF community and staff are really looking forward to being at AgentCon and MCPCon in North America. I think it’s in San Jose, October 22nd and 23rd. Angie, I know you’re opening, yeah, you’re opening the conference, right?

Angie Jones (13:48)
I am. We are so excited about this. We just finished AGNTCon, MCPCon in Europe. And Sally, let me tell you, it was electric. and I’m not just saying that because we’re throwing the conference. It really was one of the best conferences I’ve ever been to. And I’ve been to hundreds of conferences. Every… I couldn’t take two steps without someone stopping me and just saying thank you for this. This is incredible. And anyone can go on social media and see the outpouring of testimonials. Like people absolutely loved it because all of this is new, right? And we’re all trying to figure out some of the same challenges. And you shouldn’t do that alone, right? And so now you have like hundreds of companies, like we’re expecting thousands of people at the North America events. You have all of these companies there that are sharing their stories of how they solved this challenge, what they built, you know, to get ahead, right? and it’s just really refreshing. All of the content is really high-level content, like we take that very seriously. You won’t get a whole bunch of just kind of theoretical demos there. It’s not that kind of party. It’s like, no, here’s, you know, here’s what we are actually doing. Here’s what worked for us. Here’s what didn’t work for us, which I think is also important to know.

Sally Cooper (15:18)
Yeah, that’s such a good point. And yeah, I’ve been following along on social media and I’ve seen images of everybody in the room and you could feel it, it’s palpable. You could feel the electric energy of that. So I’m sure North America is gonna have the same. And I’m also excited because our community’s there. We’ve got Sarah Evans, who’s part of our AI / ML Security Working Group at the OpenSSF. And she’s gonna be with CRob talking about Agentic AI.

In the supply chain. And not only that, but they’re gonna make it fun. They’re gonna do a Candyland game or kind of like Candyland. Yeah, it’s a board game. And so they’re gonna just walk people through lessons from the software supply chain security and then apply that to Agentic AI, which is gonna be super fun. That’s one.

And then on October 23rd, we got Sarah again, Jay White and Frederick Kautz and Laura Guzzelli. And so Jay, Sarah, and Frederick are all part of the AIML security working group. And Laura is our new AI security architect here at the OpenSSF. And they’re doing a workshop on secure agentic framework for Agentic AI. Participants get to do a hands-on experience finding and addressing security threats. And there’ll be examples from MCP and Kubernetes in that one.

So I feel like this is where this conversation kind of becomes tangible. People across our communities will get together and put this into practice.

Angie Jones (16:48)
that that’s amazing I know that’s gonna be a hit security is always a top concern at these events and people want to know like you know how how do we ensure that what we’re building is actually secure and so I I already know this is gonna these those sessions will be amazing.

Sally Cooper (17:09)
Yeah, and I’m really looking forward to you opening it up. This is gonna be amazing. okay. Before I let you go, Angie, we have a little tradition here on What’s in the SOSS. It’s called rapid, rapid fire. So if you’re ready, and there’s no overthinking, I mean, just first thing that comes to your mind, I’m gonna ask you questions and you give me your answer. are you ready for rapid, rapid fire?

Angie Jones (17:32)
I’m ready. Okay.

Sally Cooper (17:34)
Favorite open source project?

Angie Jones (17:38)
Goose.

Sally Cooper (17:40)
Love it. Favorite open source mascot.

Angie Jones (17:46)
Ooh. wow, that’s a tough one. I’m gonna just go with the GitHub. Octa Octa Octo. I don’t know what the rest of it is. Octo something, but I like that one.

Sally Cooper (18:01)
Yeah, I’ve seen that one. Super cute. Okay.

Angie Jones (18:06)
Octocat!

Sally Cooper (18:01)
Yep. Octocat. Okay, most underrated developer skills.

Angie Jones (18:15)
Ooh, I I think communication and that’s becoming even more prevalent now that you need to communicate with these agents.

Sally Cooper (18:27)
Mm, good one. Okay, AI agents. Exciting, terrifying, or both?

Angie Jones (18:35)
Exciting. Yeah.

Sally Cooper (18:36)
Yeah. Amazing. Well, thank you. Thank you so much, Angie. It’s been such a pleasure having you on the podcast. And for everyone listening, make sure to join us at AgentCon and MCPCon North America October 22nd and 23rd in San Jose. I will put the registration link in the episode notes. And yeah, stop by and say hi. I think we’ll have a booth there we’re sponsoring. And until next time, stay secure, happy open sourcing. And that’s a wrap.

Angie Jones (19:07)
Thanks so much, Sally.

Announcer (19:13)
Like what you’re hearing? Be sure to subscribe to What’s in the SOSS on Spotify, Apple Podcasts, Antennapod, Pocketcast, or wherever you get your podcasts. There’s a lot going on with the OpenSSF and many ways to stay on top of it all. Check out the newsletter for open source news, upcoming events, and other happenings. Go to opensf.org/newsletter to subscribe. Connect with us on LinkedIn for the most up-to-date OpenSSF news and insight.

And be a part of the OpenSSF community at openssf.org/getinvolved. Thanks for listening, and we’ll talk to you next time on What’s in the SOSS.