Add notebook for MLflow Tracing

[TomeHirata]

trafficstars

This PR introduces a cookbook about MLflow's observability integration with google-genai Python ADK.

Resolve #432

[google-cla[bot]]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[TomeHirata]

Hi, @markmcd. We've integrated with google-genai and would love to add this cookbook. Could you help review this PR?

[markmcd]

Yep - happy to help! Would you be able to sign the CLA before we start?

[TomeHirata]

@markmcd I've signed the CLA, and the pipeline has passed.

[TomeHirata]

@markmcd Sorry to ping you again, could you please review this PR?

[Giom-V]

@TomeHirata Thanks for the PR. I've done a first pass on it, but I will also ask @markmcd to have a look.

I think the main feedback is that it should be clear how to access the dashboard when running on colab.

I also think it would be nice to write a quick README.md in the MLflow folder to introduce the tool and link to the main documentation pages.

[TomeHirata]

Hi, @Giom-V. Thank you for reviewing the cookbook. Regarding the dashboard (MLflow tracking server) on Colab, the local tracking server probably won't work in Colab, so users need to have a remote tracking server with one of several options displayed here. Do you think we should include the guide for running a remote tracking server? and if so, which environment do you prefer?

[Giom-V]

Do you think we should include the guide for running a remote tracking server? and if so, which environment do you prefer?

I think you should. As for which one, I honestly have no preference so I'd say the one you prefer or that is the easiest to set-up. If you really have no preferences, maybe kubernetes as it's our solution, but don't feel pressure to choose this one.

[TomeHirata]

Hi, @Giom-V. I've addressed your comments. Can you take another look?

[TomeHirata]

Hi, @markmcd @Giom-V. Sorry for pining you again. This tutorial is important for both Gemini and MLflow to enable users to increase the observability of their Gemini workload. I'd appreciate it if you could start reviewing this PR. I've confirmed that this notebook works well in the colab environment.

[Giom-V]

Hello @TomeHirata. Sorry for the delay, between Next, IO and the May public holidays in France, I'm way behind reviewing pull requests. I promise, I'll do it on Monday.

[Giom-V]

@TomeHirata Sorry again for the delay. Everything is ok, I just need one last ask: Can you add a "community contribution" badge, like we have in https://github.com/Giom-V/gemini-api-cookbook/blob/main/examples/Virtual_Try_On.ipynb in your notebook so that everybody sees you're the one maintaining it?

[TomeHirata]

Hi, @Giom-V. Thank you for the review. I've added the badge to the notebook!

[Giom-V]

Approved and merged. You're the last one before the IO freeze.