cauchyk

Hi all,

I’m hacking on new features for the ClickHouse native client and wanted the same “just call the model” ergonomics JavaScript and Python now enjoy. It didn’t exist for modern C++, so I wrote one.

ai‑sdk‑cpp (Apache‑2.0) gives you:

- Unified calls to OpenAI (GPT‑4o) and Anthropic (Claude 3.5) with a single C++20 API. - Streaming, multi‑turn chat, error handling—all std::optional/std::variant, no macros. - Tool calling (function‑calling) so the model can hit real APIs; sync or async, runs in parallel.

The tricky bit: C++ still lacks real reflection, so mapping plain functions → JSON schemas isn’t as automatic as, say, TypeScript decorators. I’d love fresh eyes on that part. Try the examples and tell me where it feels clunky. This is inspired by Vercel's AI SDK [1], and litellm [2].

Repo live here: https://github.com/ClickHouse/ai-sdk-cpp, feedback welcome!

[1] https://github.com/vercel/ai [2] https://github.com/BerriAI/litellm

Reference: https://x.com/saisrirampur/status/1824694191537959159 Parent Tweet: https://x.com/craigkerstiens/status/1824114371737616794

Change Data Capture(CDC) is hard, it has 100s of edge cases / failure points. At PeerDB (https://www.peerdb.io/) , instead of focusing on multiple sources we just focused on Postgres. This helped us ensure that we gave enough care to iron out as many edge cases as possible. We were also able to implement a bunch of Postgres native performance and reliability optimizations. Our engineering blog https://blog.peerdb.io/ more on the optimizations and how we ironed out edge cases.

Pipeline failures have been rare these days, and so far none of the source databases were affected due to load. Also, most of our customers are in the shorter tail, i.e., data sizes over 300-400GB to 15-20TB. This helped battle test the product and make seamless for the long tail.

However, I don’t think CDC is a solved problem, as Postgres is full of mysteries and it keeps evolving. We need to continue polishing the experience and evolve along with Postgres!

TL;DR specialized CDC tools that focus on a single (or limited) database are reliable ways to provide a solid CDC experience.

I'm looking for github repositories of applications. Note, I do not care how many stars/forks a repo has or how useful the application itself is. I'm looking for well structured tested code that can serve as a reference to write a macOS application.