llmware VS fast-data-dev

I couldn't agree more. You should check out LLMWare's SLIM agents (https://github.com/llmware-ai/llmware/tree/main/examples/SLI...). It's focusing on pretty much exactly this and chaining multiple local LLMs together.

A really good topic that ties in with this is the need for deterministic sampling (I may have the terminology a bit incorrect) depending on what the model is indended for. The LLMWare team did a good 2 part video on this here as well (https://www.youtube.com/watch?v=7oMTGhSKuNY)

I think dedicated miniture LLMs are the way forward.

Disclaimer - Not affiliated with them in any way, just think it's a really cool project.

I've been building upon the LLMWare project - https://github.com/llmware-ai/llmware - for the past 3 months. The ability to run these models locally on standard consumer CPUs, along with the abstraction provided to chop and change between models and different processes is really cool.

I think these SLIM models are the start of something powerful for automating internal business processes and enhancing the use case of LLMs. Still kinda blows my mind that this is all running on my 3900X and also runs on a bog standard Hetzner server with no GPU.

16. LLMWare by Ai Bloks | Github | tutorial

I liked https://lenses.io/ Lots of capabilities but it's not free as I know.

Lenses

For Kafka Dev I use https://github.com/lensesio/fast-data-dev

Currently I'm using Lenses:https://lenses.io/ as UI tool, but while turning on kafka-start-server server.properties and launch the UI on localhost, it failed to connect:

Oh thats sad to hear ... lenses.io is so powerful I am not sure how I would have gotten by to this point without it!

In Addition to the more critical replys from others, why Kafka still makes sense imo: - you could organist everything with custom pipeline/services at your scale. The cost of maintaining those and keep the technology up to date is growing exponentially with every new Service. Kafka offers ansinge Plattform with a couple components which need to be kept up to date. - deploying, managing and monitoring those service is also a big task and does not come for free. - unless your using Kubernetes underneath, scaling is not easy with services. - peer to peer communications of microservices should be avoided because your building a „deathstar“-like architecture where everything is linked and effectively you’ll end up with a distributed monolith. A message broker or message queue helps to avoid this, Kafka is a good tool for that. - I heard that those aws Kafka replacements are not 100% Kafka as one might expect, you should be careful there. But confluent or other providers probably offer hosted solutions on aws too. - https://lenses.io/ helped us a lot when we started with Kafka to see how everything works together, create stream processor’s with sql and easily deploy Kafka connect workers. The tool costs money though but is worth it in the end in my opinion. (Disclaimer: I don’t work for confluent or lenses, we’re running open source Kafka with strimzi and lenses)

https://github.com/lensesio/fast-data-dev is a really cool option for a Kafka development environment based on docker and/or kubernetes.

Yes: - https://lenses.io/ (use the Lenses box, it’s free) - https://github.com/lensesio/fast-data-dev - https://github.com/provectus/kafka-ui