Show HN: LangChain, but for Software Engineers (opens in new tab)

I have built many AI agents, and their development always felt weird because all frameworks were mangling together everything from data stores to data processing, which felt super weird to me as a software engineer.

I built Flashlearn which is completely JSON driven(JSON in, JSON, out) and it only supports interacting with LLMs everything else you do exactly like you would in normal software development.

It builds on the single-responsibility principle - one action one call to LLM.

It has a fit/predict pattern for learning to complete new tasks with LLMs.

In addition to this, there is also concurrency support 1000 calls/min.

You can simply store your JSON artifacts and call LLMs for any system, track them edit them etc.

Oh and it always returns structured responses that are always consistent and you can simply use them in downstream tasks:)