“So for now we probably should use it only for tasks where facts are not important, such as writing letters of recommendation and formulating government policy.”
And for some reason I used to think security patches get back ported to all the supported versions and by not upgrading I was only missing out on new features.
Thanks, I’m realizing that now. It helps that you’re emphasizing the need to stay up to date.
I don’t upgrade to the latest version when it comes out thinking it may not be stable enough yet. And then I remember about it when I’m about to start working or in the flow. I know, silly excuses!
Didn’t try the password reset until you mentioned. Thanks, that worked.
Google did send me two Security alerts (one for each laptop) when I tried signing in yesterday with my old pwd. So they must have reset my password or something?
In any case, lesson learned: never connect to an unsecured Wi-Fi again! (I rarely do, but I was at this conference last week trying to demo Appomate AI, and was wanting it to be as snappy as possible. Bad decision!)
Good to know! Thanks for sharing your workflow.
To answer your question, I just think when you're scrolling through git log, seeing a message in your own words has a much better chance of bringing back all that went behind that commit. But I agree that's very subjective.
> It's bizarre: Standardize the API via individually wrapped endpoints. WTF?
Exactly!
> Yes, after spending more time trying to decypher the LangChain docs than it would take to roll-my-own, everything I've done so far has involved rolling-my-own.
Looks like quite a few are doing the same, that is, rolling their own. Wonder if there needs to be a way to do it where it's standardized but still feels like rolling your own.
> I still think I might need to work with LangChain however: I'm afraid it's going to be what employers will be looking for.
Yeah workplaces are so different than working on your own projects. Been there! Might be you can get started that way, and slowly help people "see the light"!
I feel the same! Just look at all the separate classes for different model providers for tackling minor input/output differences.
As for the alternatives, I think there are 3 things to think about:
1. LLM call itself
2. Orchestration of calls & monitoring/input/output
3. Integrations with external services
For #1, if we were to limit the discussion to model providers with REST APIs (OpenAI, Anthropic, Google, Cohere, Groq, Together, ElevenLabs all have them), I think it would be much simpler to build a simple wrapper directly on top of fetch or equivalent APIs and even skip the SDKs.
For #2, don't think LangChain is a better solution than alternatives like state charts.
#3 is where LangChain might offer some value, but it's pretty thin.
“So for now we probably should use it only for tasks where facts are not important, such as writing letters of recommendation and formulating government policy.”
:-)