EmadKhair

I actually agree with what you said! the subscriptions we are paying today is only a tiny fraction of what it actually cost AI companies like Claude and Codex. either it will shift into running those agents in your own setup or pay an absurred number. anyways as you said the pendulum will swing back for sure!

My siblings and I are building Octri, an all-in-one platform for API teams.

https://octri.dev

You give it an OpenAPI, Postman, or AsyncAPI spec, and it generates and manages:

- Editable API docs with an API playground, RAG-based chat, custom components, and other interactive features.

- Idiomatic SDKs in 10 languages. We’ve been building these alongside developers experienced in each language rather than relying on a single generic codegen template.

The part we’ve spent a lot of time on is customization. There’s a visual editor with a real-time code preview where you can change things like: - client initialization patterns - positional args vs DTOs vs kwargs - HTTP client/library - tree-shakability - method structure and naming - and quite a bit more

The goal is for generated SDKs to feel like something you would have written by hand rather than obviously generated code.

- API + SDK monitoring that maps telemetry back to your API spec. It can identify slow SDK methods and API paths, track uptime/status announcements, capture errors and stack traces, etc. We’ve also built PII filtering into the telemetry pipeline with GDPR considerations in mind.

- Spec Audit, which we’re planning to open source. It scores an OpenAPI spec, explains problems it finds, and can automatically fix a number of them.

A big reason we started working on this is that API tooling in this space often feels optimized for enterprise buyers rather than the developers actually using it. We also kept running into rough UX around editing docs, customizing generated SDKs, and specs failing on first import because of small incompatibilities.

We want Octri to have a much lower barrier to entry: affordable enough for smaller teams, highly customizable, and able to improve/fix inputs rather than just failing when a spec isn’t perfect.

We’re still in beta and looking for a few design partners who have real APIs/SDKs they’d be willing to run through it. In return, we’d offer favorable long-term pricing for early adopters.

I’d also genuinely appreciate feedback from people here who maintain public APIs or SDKs: what part of your API tooling workflow is currently the most painful?

cool idea but no one will want to subscribe without validating that it works first. maybe add some tokens to test it out and give out quality results (limit the amount of results if you need to) to encourage users to upgrade and get more?

I think the idea is great! just one thing as a user I was kinda lost with the "concept" posts. maybe you should have less of those "concepts" and prioritize user made trips to show on the top? that way it would encourage people to check out real trips and consider them rather than keep creating and flooding the site with no trips actually filled.

Love this! At an era where AI is automating everything, it's good to see a product that reminds us of the way it was before computers and AI got involved. (I need to work on my handwriting! haven't used a pen in ages! X))