README.md

kbase-chat-assistant

KBase Research Assistant is powered by multiple Language Model (LLM)-based Agents designed to interact conversationally with users. This Assistant will serve as a guide, facilitating users' navigation within the KBase platform.

Uses Poetry to handle requirements

Installation

1. Setup a conda environment
2. Run poetry install

Deployment on SPIN at NERSC

TODO. For now, see the SpinUp docs and tutorials. You'll be making a pretty simple deployment of a single node, with a single service, with a single ingress to get this up and running.

Updating a deployment on SPIN at NERSC

1. Build the new docker image

docker build --platform="linux/amd64" -t registry.nersc.gov/kbase/kbase-chat-assistant .

2. Push to the registry. You might need to login with your NERSC id, pw+2fa

docker push registry.nersc.gov/kbase/kbase-chat-assistant

3. Log in to https://rancher2.spin.nersc.gov
4. Click on the Development server
5. On the left side find "Workloads". Click that, then "Deployments" underneath.
6. Find the chat assistant (namespace: knowledge-engine, container: chat-assistant)
7. On the far right, hit the 3-dots menu, then Redeploy
8. It can take a while for the container to get the memo that it should shut down, so just let it go.

TODOs

1. Add the Knowledge Graph
2. User interface