Project Description

Over the years, our bugzilla database has grown up in size, becoming a very valuable source of truth for most support and development cases; still searching for specific items is quite tricky and the results do not always match the expectations.

What about feeding a Maching Learning platform with the Bugzilla Database, in order to be able to query it through AI interface? Wouldn't it be nice/convenient to ask to AI: "Gimme hints about this kernel dump!" or "What is the root cause of this stack trace?"

It is the age of choice in the end, isn't it?

Goal for this Hackweek

For this Hackweek, the focus is to trigger a discussion on the following non-exhaustive list:

  • What are the boundaries to be set when considering such an approach (legal, ethical, technological, whatever)
  • How much of the Bugzilla DB can be used for feeding ML? ( can we use customer's data? what about partner's data?)
  • Find out an open source ML solution fitting our needs;
  • Find out some hardware where the solution can be eventually run on.

Anyone interested can join the discussion on the open Slack channel #discuss-bugzilla-ai

Resources

[1] https://blog.opensource.org/towards-a-definition-of-open-artificial-intelligence-first-meeting-recap/

Looking for hackers with the skills:

ai machinelearning machine-learning bugzilla support

This project is part of:

Hack Week 23

Activity

  • 8 months ago: wfrisch liked this project.
  • 8 months ago: jmodak liked this project.
  • 9 months ago: cxiong liked this project.
  • 9 months ago: lthadeus liked this project.
  • 9 months ago: ygutierrez liked this project.
  • 9 months ago: paolodepa added keyword "machine-learning" to this project.
  • 9 months ago: paolodepa added keyword "bugzilla" to this project.
  • 9 months ago: paolodepa added keyword "support" to this project.
  • 9 months ago: paolodepa added keyword "ai" to this project.
  • 9 months ago: paolodepa added keyword "machinelearning" to this project.
  • 9 months ago: paolodepa originated this project.

  • Comments

    • paolodepa
      8 months ago by paolodepa | Reply

      Preliminary findings: talking to Amartya Chakraborty, who works to the Rancher AI project (https://github.com/rancher/opni), it seems that their framework can be attached to a Bugzilla instance for machine learning and pobably this will be explorated in the future

    • paolodepa
      8 months ago by paolodepa | Reply

      Preliminary finding: the Mozilla foundation is actively working on https://github.com/mozilla/bugbug, coming with very promising features: it's worth to try to setup an instance and feed it using our Bugzilla data.

    • paolodepa
      8 months ago by paolodepa | Reply

      Suspended due to flu: feel free to take-over

    • paolodepa
      8 months ago by paolodepa | Reply

      Postponed to upcoming hackweeks

    Similar Projects

    COOTWbot by ngetahun

    [comment]: # (Please use the project descriptio...


    A quantum physics experiment puzzle (designed with Google's CP-SAT solver) by moio

    [![link to video player demoing the result](htt...


    Package MONAI Machine Learning Models for Medical Applications by jordimassaguerpla

    Project Description

    MONAI Deploy aims to ...


    Create language model based on Uyuni documentation by nadvornik

    [comment]: # (Please use the project descriptio...


    Implement Generative AI for SUSE Asset management (using Amazon Bedrock) by mpiala

    [comment]: # (Please use the project descriptio...


    Bugzilla mobile client by paolodepa

    Project Description

    _DISCLAIMER: I'm **no...