Derek Thompson of Ambassador came by for a discussion about code generation and artificial intelligence. I had began engaging with Derek's team about the importance of maintaining OpenAPI specs while generating code on Bluesky, and they agreed to have Derek come and share more of his expertise with me on the subject. I was very pleased to hear that first, they weren't focused on client SDKs, and they are generating server stubs, mock servers, and all the details that go with that. I was also very satisfied to hear of their embrace of OpenAPI, but also overlays and Arazzo. I enjoyed Derek's expertise in both API but also AI, and his focus on the developer, as well as respect for OpenAPI.

I sat down today with Marc Laventure of Scalar to talk about their API client and documentation offerings, and how they see OpenAPI as a specification. Marc's taken a slightly different approach than other API clients solutions and centering their solutions on just OpenAPI, rather than creating another spec. I feel like it is a sensible approach and can help reduce the complexity across API operations. Marc also shared his vision for their business model and where they will make money, and how they intend to balance that with their very modular and local open-source solutions--which is something I can get behind.

Juan Cruz Viotti of Sourcemeta came by for a conversation about the schema registry solution he had recently built. Juan has gone deep into the JSON Schema rabbit hole, deeper than anyone else I've ever met. His validation tooling and schema registry are providing a new foundation for enterprises to get their schema house in order. This is something Juan and I both agree is the single most foundational thing you can invest in when it comes to API governance. I learn so much everytime I talk to Juan and explore his work, and we will be working together more around the evolution of his schema registry, and API Evangelist will be leaning more into using the JSON Schema educational resources he's contributed to at https://www.learnjsonschema.com/2020-12/.

Frank Kilcommins of SmartBear came by to talk about the OpenAPI Arazzo specification. He shared the history how it came to be and how it is changing the conversation inside and outside the OpenAPI Initiative (OAI). Franks done amazing work to move the API workflows specification over the finish line, but also keep it evolving and moving forward afterwards. The specification has renewed energy with API service providers who are integrating the spec into their API guided walkthroughs, SDK generation, and testing. I am a big fan of Frank's work, and wish we had 10 more of him working at the OpenAPI, so if you are feeling like you have some spare cycles, I recommend you rolling up your sleeves and getting involved. Thanks for coming by and sharing Frank!!

Adron Hall came by for a conversation about trains. Adron and I share a love of transit and trains, who also make their living slinging APIs. Adron had recommended I read Nothing Like It in the World by Stephen E Ambrose and Empire Express by David Haward Bain back a couple of years ago, which I finally did. Then things kept happening with AI and the current presidency that kept reminding me of the stories in these books and I kepting thinking it would be fun to have Adron come by to share his view of things. I won't spoil any of his great stories, but both books and Adron's view of them, as well as his ability to apply in this moment is gold. Adron is a great storyteller and he gets the technology, but he also the business and politics of technology in a way few do, making for a very entertaining 30 minutes.

Tech storyteller and journalist Jennifer Riggins came by for a conversation about why nothing has changed in the world of APIs for the last fifteen years. Jennifer and I go all the way back to the beginning, and she has seen all the same things I've seen. We complained a bit about the impacts of AI, but really that most of the lack of change has been a business thing and not really just technical. I like Jennifer's focus on developers in this moment, and she got me thinking about the relationship between the legacy API management we both have lived through and this new platform reality we operate within. I always enjoy sitting down with Jennifer, and it has been way too long, appreciate you coming by to share some wisdom around why things aren't as progressive as they could be in the API space.

Quobix from from Princess Beef Heavy Industries came by again, this time to talk about OpenAPI Doctor and libopenapi at it's core. I knew "The Doctor" was a modular OpenAPI editor w/ governance built-in, but I didn't realize how deeply modularity is baked into it's design. This modularity allows Quobix to stitch together Vacuum, libopenapi, OpenAPI changes, wiretap, and other Princess Beef Heavy Industries solutions, but also potentially other 3rd party solutions. This modularity in API hubs, dashboards, and tools is fast becoming a common themes in conversations that I am having, and something I am going to work to leverage more as part of my work. This episode went 30 minutes because I wanted to do two separate episoides, 1) OpenAPI, 2) libopenapi, but the stories are so entertwhined we just did it back to back.

Daniel Kocot came by again to talk about the lines between our private and public APIs. We had been talking about internal, first-party, and third-party APIs back and forth on LinkedIn, and I recommended that he come by and we will have a conversation. He helped me realize that my separation between the layers was more about access, control, and velocity over just about being inside or outside the firewall and DMZ. Daniel sees this for what it is--it is about people. Lines of business, teams, tribes, and partners. Daniel has that very outside-in view of API operations, but Codecentric is being brought inside to solve the big problems for their customers. Talking with Daniel makes me realize how important it is to have these conversations about our APIs and portals, because not everyone has a handle on the nuance of what is happening at the edge of our enterprises.

Adrian Machado from Zuplo came by for a talk on API rate limiting, but not just the technical bits. Zuplo is doing a good job at bridging the business and technical aspects of rate limiting at the API gateway layer, and we had a compelling discussion around how planning your rate limits will shape your overall API business plan. Adrian provided some compelling thoughts around the need for standardization of API rate limits and how we approach the API access plans and policies we are implementing in regions around the globe. After the gateway layer Adrian shared more about the Zuplo ethos of API documentation should be free with their Zudoku offering, something I can definitely get behind. Thanks for the great talk Adrian, come by antyime.

I sat down with Vincent Biret, Microsoft Graph SDKs Principal Software Developer at Microsoft to talk about all things integration. I am interested in the intersection of APIs, specifications, and integrations where Vincent operates. Vincent shared his view of important role that OpenAPI plays in taming the Microsoft Graph API landscape, but specifically how it gets applied as part of producer or consumer driven SDK generation using Kiota. Vincent shared all kinds of experience and wisdom at this intersection, but he also shared a pragmatic view of the role APIs, specs, and integrations play when it comes to training LLMs, but also feeding agents deployed on top of those LLMs with the real-time resources and capabilities they will need. It was an illuminating conversation and one I learned a lot, and will be crafting more stories about over the coming weeks.