Learn how to integrate Elixir and Phoenix Liveview with LLMs by ninopaparo

Description

Learn how to integrate Elixir and Phoenix Liveview with LLMs by building an application that can provide answers to user queries based on a corpus of custom-trained data.

Goals

Develop an Elixir application via the Phoenix framework that:

• Employs Retrieval Augmented Generation (RAG) techniques
• Supports the integration and utilization of various Large Language Models (LLMs).
• Is designed with extensibility and adaptability in mind to accommodate future enhancements and modifications.

Resources

• https://elixir-lang.org/
• https://www.phoenixframework.org/
• https://github.com/elixir-nx/bumblebee
• https://ollama.com/

---

Save pytorch models in OCI registries by jguilhermevanz

Description

A prerequisite for running applications in a cloud environment is the presence of a container registry. Another common scenario is users performing machine learning workloads in such environments. However, these types of workloads require dedicated infrastructure to run properly. We can leverage these two facts to help users save resources by storing their machine learning models in OCI registries, similar to how we handle some WebAssembly modules. This approach will save users the resources typically required for a machine learning model repository for the applications they need to run.

Goals

Allow PyTorch users to save and load machine learning models in OCI registries.

Resources

---

Gen-AI chatbots and test-automation of generated responses by mdati

Description

Start experimenting the generative SUSE-AI chat bot, asking questions on different areas of knowledge or science and possibly analyze the quality of the LLM model response, specific and comparative, checking the answers provided by different LLM models to a same query, using proper quality metrics or tools or methodologies.

Try to define basic guidelines and requirements for quality test automation of AI-generated responses.

First approach of investigation can be based on manual testing: methodologies, findings and data can be useful then to organize valid automated testing.

Goals

• Identify criteria and measuring scales for assessment of a text content.
• Define quality of an answer/text based on defined criteria .
• Identify some knowledge sectors and a proper list of problems/questions per sector.
• Manually run query session and apply evaluation criteria to answers.
• Draft requirements for test automation of AI answers.

Resources

• Announcement of SUSE-AI for Hack Week in Slack
• Openplatform and related 3 LLM models gemma:2b, llama3.1:8b, qwen2.5-coder:3b.

Notes

• Foundation models (FMs):
  are large deep learning neural networks, trained on massive datasets, that have changed the way data scientists approach machine learning (ML). Rather than develop artificial intelligence (AI) from scratch, data scientists use a foundation model as a starting point to develop ML models that power new applications more quickly and cost-effectively.

• Large language models (LLMs):
  are a category of foundation models pre-trained on immense amounts of data acquiring abilities by learning statistical relationships from vast amounts of text during a self- and semi-supervised training process, making them capable of understanding and generating natural language and other types of content , to perform a wide range of tasks.
  LLMs can be used for generative AI (artificial intelligence) to produce content based on input prompts in human language.

Validation of a AI-generated answer is not an easy task to perform, as manually as automated.
An LLM answer text shall contain a given level of informations: correcness, completeness, reasoning description etc.
We shall rely in properly applicable and measurable criteria of validation to get an assessment in a limited amount of time and resources.

---

ghostwrAIter - a local AI assisted tool for helping with support cases by paolodepa

Description

This project is meant to fight the loneliness of the support team members, providing them an AI assistant (hopefully) capable of scraping supportconfigs in a RAG fashion, trying to answer specific questions.

Goals

• Setup an Ollama backend, spinning one (or more??) code-focused LLMs selected by license, performance and quality of the results between:
  • deepseek-coder-v2
  • dolphin-mistral
  • starcoder2
  • (...others??)
    
• Setup a Web UI for it, choosing an easily extensible and customizable option between:
  • openwebui
  • localAI
  • lobachat
  • (...others??)
    
• Extend the solution in order to be able to:
  • Add ZIU/Concord shared folders to its RAG context
  • Add BZ cases, splitted in comments to its RAG context
    • A plus would be to login using the IDP portal to ghostwrAIter itself and use the same credentials to query BZ
  • Add specific packages picking them from IBS repos
    • A plus would be to login using the IDP portal to ghostwrAIter itself and use the same credentials to query IBS
    • A plus would be to desume the packages of interest and the right channel and version to be picked from the added BZ cases

---

Use local/private LLM for semantic knowledge search by digitaltomm

Description

Use a local LLM, based on SUSE AI (ollama, openwebui) to power geeko search (public instance: https://geeko.port0.org/).

Goals

Build a SUSE internal instance of https://geeko.port0.org/ that can operate on internal resources, crawling confluence.suse.com, gitlab.suse.de, etc.

Resources

Repo: https://github.com/digitaltom/semantic-knowledge-search

Public instance: https://geeko.port0.org/

Results

Internal instance:

I have an internal test instance running which has indexed a couple of internal wiki pages from the SCC team. It's using the ollama (llama3.1:8b) backend of suse-ai.openplatform.suse.com to create embedding vectors for indexed resources and to create a chat response. The semantic search for documents is done with a vector search inside of sqlite, using sqlite-vec.

---

Symbol Relations by hli

Description

There are tools to build function call graphs based on parsing source code, for example, cscope.

This project aims to achieve a similar goal by directly parsing the disasembly (i.e. objdump) of a compiled binary. The assembly code is what the CPU sees, therefore more "direct". This may be useful in certain scenarios, such as gdb/crash debugging.

Detailed description and Demos can be found in the README file:

Supports x86 for now (because my customers only use x86 machines), but support for other architectures can be added easily.

Tested with python3.6

Goals

Any comments are welcome.

Resources

https://github.com/lhb-cafe/SymbolRelations

symrellib.py: mplements the symbol relation graph and the disassembly parser

symrel_tracer*.py: implements tracing (-t option)

symrel.py: "cli parser"

---

Selenium with Python by xguo

Description

Try to create test case about Selenium base on Python

Goals

• Knowledge about Selenium with Python
• Create new test case about Selenium

Resources

https://selenium-python.readthedocs.io/ https://www.selenium.dev/

---

Enhance UV openQA helper script by mdonis

Description

A couple months ago an UV openQA helper script was created to help/automate the searching phase inside openQA for a given MU to test. The script searches inside all our openQA job groups (qam-sle) related with a given MU and generates an output suitable to add (copy & paste) inside the update log.

This is still a WIP and could use some enhancements.

Goals

• Move script from bash to python: this would be useful in case we want to include this into MTUI in the future. The script will be separate from MTUI for now. The idea is to have this as a CLI tool using the click library or something similar.
• Add option to look for jobs in other sections inside aggregated updates: right now, when looking for regression tests under aggregated updates for a given MU, the script only looks inside the Core MU job group. This is where most of the regression tests we need are located, but some MUs have their regression tests under the YaST/Containers/Security MU job groups. We should keep the Core MU group as a default, but add an option to be able to look into other job groups under aggregated updates.
• Remove the -a option: this option is used to indicate the update ID and is mandatory right now. This is a bit weird and goes against posix stardards. It was developed this way in order to avoid using positional parameters. This problem should be fixed if we move the script to python.

Some other ideas to consider:

• Look into the QAM dashboard API. This has more info on each MU, could use this to link general openQA build results, whether the related RR is approved or not, etc
• Make it easier to see if there's regression tests for a package in an openQA test build. Check if there's a possibility to search for tests that have the package name in them inside each testsuite.
• Unit testing?

More ideas TBD

Resources

https://github.com/os-autoinst/scripts/blob/master/openqa-search-maintenance-core-jobs

https://confluence.suse.com/display/maintenanceqa/Guide+on+how+to+test+Updates

Post-Hackweek update

All major features were implemented. Unit tests are still in progress, and project will be moved to the SUSE github org once everything's done. https://github.com/mjdonis/oqa-search

---

Small healthcheck tool for Longhorn by mbrookhuis

Project Description

We have often problems (e.g. pods not starting) that are related to PVCs not running, cluster (nodes) not all up or deployments not running or completely running. This all prevents administration activities. Having something that can regular be run to validate the status of the cluster would be helpful, and not as of today do a lot of manual tasks.

As addition (read enough time), we could add changing reservation, adding new disks, etc. --> This didn't made it. But the scripts can easily be adopted.

This tool would decrease troubleshooting time, giving admins rights to the rancher GUI and could be used in automation.

Goal for this Hackweek

At the end we should have a small python tool that is doing a (very) basic health check on nodes, deployments and PVCs. First attempt was to make it in golang, but that was taking to much time.

Overview

This tool will run a simple healthcheck on a kubernetes cluster. It will perform the following actions:

• node check: This will check all nodes, and display the status and the k3s version. If the status of the nodes is not "Ready" (this should be only reported), the cluster will be reported as having problems

• deployment check: This check will list all deployments, and display the number of expected replicas and the used replica. If there are unused replicas this will be displayed. The cluster will be reported as having problems.

• pvc check: This check will list of all pvc's, and display the status and the robustness. If the robustness is not "Healthy", the cluster will be reported as having problems.

If there is a problem registered in the checks, there will be a warning that the cluster is not healthy and the program will exit with 1.

The script has 1 mandatory parameter and that is the kubeconf of the cluster or of a node off the cluster.

The code is writen for Python 3.11, but will also work on 3.6 (the default with SLES15.x). There is a venv present that will contain all needed packages. Also, the script can be run on the cluster itself or any other linux server.

Installation

To install this project, perform the following steps:

• Create the directory /opt/k8s-check

mkdir /opt/k8s-check

• Copy all the file to this directory and make the following changes:

chmod +x k8s-check.py

---