https://github.com/crdoconnor/tmg-test
TMG solution
https://github.com/crdoconnor/tmg-test
Last synced: over 1 year ago
JSON representation
TMG solution
- Host: GitHub
- URL: https://github.com/crdoconnor/tmg-test
- Owner: crdoconnor
- Created: 2024-09-23T22:45:43.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-09-24T12:36:24.000Z (almost 2 years ago)
- Last Synced: 2025-01-23T02:11:26.236Z (over 1 year ago)
- Language: Python
- Size: 10.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# TMG test
To run:
* Check out repository
* Put OPENAI_API_KEY=<< openai key >> in .env file
* make build
* make run
To test:
* make test
## Deploying in the cloud
* Use poetry / pipenv to fix requirements.
* Don't bake the secrets into the docker container.
* Push docker container to registry
* Create secret in secret storage in cloud
* Create service manifest (e.g. in kubernetes) which will run container as a service and inject the secret as an environment variable
## Testing
* Already has one test - end to end (might be flaky).
* Inject the LLM Runner as a dependency in fastapi, swap with fake to run dependency-less unit tests.
* For testing the LLM build up a test set of recipes and expected tags and evaluate response.