7 ms·
Most ML applications are just request routers
- rossamurphy 2y agoMoving an internal ML project from "a quick demo on localhost", to "deployed in production", is hard. We think latency is one of the biggest problems. We built OneContext to solve that problem. We launched today. Would love your feedback + feature requests!
- harindirand 2y agoLooks super interesting! This could be super helpful for us. Will drop your team a note :)
- cwmdo 2y ago“simply by cutting out the network latency between the steps, OneContext reduces the pipeline execution time by 57%)” how does this fit in with barebones langchain/bedrock setup?
- the_async 2y agoSeems like a great product !
- georgespencer 2y agoAmazing! Congrats on launching. Company motto: "dumb enough to actually have attempted this already".