Self-Scaling LLM Infrastructure Powered by Rancher by ademicev0

Self-Scaling LLM Infrastructure Powered by Rancher

---

Description

The Problem

Running LLMs can get expensive and complex pretty quickly.

Today there are typically two choices:

1. Use cloud APIs like OpenAI or Anthropic. Easy to start with, but costs add up at scale.
2. Self-host everything - set up Kubernetes, figure out GPU scheduling, handle scaling, manage model serving... it's a lot of work.

What if there was a middle ground?

What if infrastructure scaled itself instead of making you scale it?

Can we use existing Rancher capabilities like CAPI, autoscaling, and GitOps to make this simpler instead of building everything from scratch?

Project Repository: github.com/alexander-demicev/llmserverless

---

What This Project Does

A key feature is hybrid deployment: requests can be routed based on complexity or privacy needs. Simple or low-sensitivity queries can use public APIs (like OpenAI), while complex or private requests are handled in-house on local infrastructure. This flexibility allows balancing cost, privacy, and performance - using cloud for routine tasks and on-premises resources for sensitive or demanding workloads.

A complete, self-scaling LLM infrastructure that:

• Scales to zero when idle (no idle costs)
• Scales up automatically when requests come in
• Adds more nodes when needed, removes them when demand drops
• Runs on any infrastructure - laptop, bare metal, or cloud

Think of it as "serverless for LLMs" - focus on building, the infrastructure handles itself.

How It Works

A combination of open source tools working together:

Flow:

• Users interact with OpenWebUI (chat interface)
• Requests go to LiteLLM Gateway
• LiteLLM routes requests to:
  • Ollama (Knative) for local model inference (auto-scales pods)
  • Or cloud APIs for fallback

---

Update M2Crypto by mcepl

There are couple of projects I work on, which need my attention and putting them to shape:

• M2Crypto

Goal for this Hackweek

• Put M2Crypto into better shape (most issues closed, all pull requests processed)
• More fun to learn jujutsu
• Play more with Gemini, how much it help (or not).
• Perhaps, also (just slightly related), help to fix vis to work with LuaJIT, particularly to make vis-lspc working.

---

Try AI training with ROCm and LoRA by bmwiedemann

Description

I want to setup a Radeon RX 9600 XT 16 GB at home with ROCm on Slowroll.

Goals

I want to test how fast AI inference can get with the GPU and if I can use LoRA to re-train an existing free model for some task.

Resources

• https://rocm.docs.amd.com/en/latest/compatibility/compatibility-matrix.html
• https://build.opensuse.org/project/show/science:GPU:ROCm
• https://src.opensuse.org/ROCm/
• https://www.suse.com/c/lora-fine-tuning-llms-for-text-classification/

Results

got inference working with llama.cpp:

export LLAMACPP_ROCM_ARCH=gfx1200
HIPCXX="$(hipconfig -l)/clang" HIP_PATH="$(hipconfig -R)" \
cmake -S . -B build -DGGML_HIP=ON -DAMDGPU_TARGETS=$LLAMACPP_ROCM_ARCH \
-DCMAKE_BUILD_TYPE=Release -DLLAMA_CURL=ON \
-Dhipblas_DIR=/usr/lib64/cmake/hipblaslt/ \
&& cmake --build build --config Release -j8
m=models/gpt-oss-20b-mxfp4.gguf
cd $P/llama.cpp && build/bin/llama-server --model $m --threads 8 --port 8005 --host 0.0.0.0 --device ROCm0 --n-gpu-layers 999

Without the --device option it faulted. Maybe because my APU also appears there?

I updated/fixed various related packages: https://src.opensuse.org/ROCm/rocm-examples/pulls/1 https://src.opensuse.org/ROCm/hipblaslt/pulls/1 SR 1320959

benchmark

I benchmarked inference with llama.cpp + gpt-oss-20b-mxfp4.gguf and ROCm offloading to a Radeon RX 9060 XT 16GB. I varied the number of layers that went to the GPU:

• 0 layers 14.49 tokens/s (8 CPU cores)
• 9 layers 17.79 tokens/s 34% VRAM
• 15 layers 22.39 tokens/s 51% VRAM
• 20 layers 27.49 tokens/s 64% VRAM
• 24 layers 41.18 tokens/s 74% VRAM
• 25+ layers 86.63 tokens/s 75% VRAM (only 200% CPU load)

So there is a significant performance-boost if the whole model fits into the GPU's VRAM.

---

MCP Trace Suite by r1chard-lyu

Description

This project plans to create an MCP Trace Suite, a system that consolidates commonly used Linux debugging tools such as bpftrace, perf, and ftrace.

The suite is implemented as an MCP Server. This architecture allows an AI agent to leverage the server to diagnose Linux issues and perform targeted system debugging by remotely executing and retrieving tracing data from these powerful tools.

• Repo: https://github.com/r1chard-lyu/systracesuite
• Demo: Slides

Goals

1. Build an MCP Server that can integrate various Linux debugging and tracing tools, including bpftrace, perf, ftrace, strace, and others, with support for future expansion of additional tools.

2. Perform testing by intentionally creating bugs or issues that impact system performance, allowing an AI agent to analyze the root cause and identify the underlying problem.

Resources

• Gemini CLI: https://geminicli.com/
• eBPF: https://ebpf.io/
• bpftrace: https://github.com/bpftrace/bpftrace/
• perf: https://perfwiki.github.io/main/
• ftrace: https://github.com/r1chard-lyu/tracium/

---

Extended private brain - RAG my own scripts and data into offline LLM AI by tjyrinki_suse

Description

For purely studying purposes, I'd like to find out if I could teach an LLM some of my own accumulated knowledge, to use it as a sort of extended brain.

I might use qwen3-coder or something similar as a starting point.

Everything would be done 100% offline without network available to the container, since I prefer to see when network is needed, and make it so it's never needed (other than initial downloads).

Goals

1. Learn something about RAG, LLM, AI.
2. Find out if everything works offline as intended.
3. As an end result have a new way to access my own existing know-how, but so that I can query the wisdom in them.
4. Be flexible to pivot in any direction, as long as there are new things learned.

Resources

To be found on the fly.

Timeline

Day 1 (of 4)

• Tried out a RAG demo, expanded on feeding it my own data
• Experimented with qwen3-coder to add a persistent chat functionality, and keeping vectors in a pickle file
• Optimizations to keep everything within context window
• Learn and add a bit of PyTest

Day 2

• More experimenting and more data
• Study ChromaDB
• Add a Web UI that works from another computer even though the container sees network is down

Day 3

• The above RAG is working well enough for demonstration purposes.
• Pivot to trying out OpenCode, configuring local Ollama qwen3-coder there, to analyze the RAG demo.
• Figured out how to configure Ollama template to be usable under OpenCode. OpenCode locally is super slow to just running qwen3-coder alone.

Day 4 (final day)

• Battle with OpenCode that was both slow and kept on piling up broken things.
• Call it success as after all the agentic AI was working locally.
• Clean up the mess left behind a bit.

Blog Post

Summarized the findings at blog post.

---

issuefs: FUSE filesystem representing issues (e.g. JIRA) for the use with AI agents code-assistants by llansky3

Description

Creating a FUSE filesystem (issuefs) that mounts issues from various ticketing systems (Github, Jira, Bugzilla, Redmine) as files to your local file system.

And why this is good idea?

• User can use favorite command line tools to view and search the tickets from various sources
• User can use AI agents capabilities from your favorite IDE or cli to ask question about the issues, project or functionality while providing relevant tickets as context without extra work.
• User can use it during development of the new features when you let the AI agent to jump start the solution. The issuefs will give the AI agent the context (AI agents just read few more files) about the bug or requested features. No need for copying and pasting issues to user prompt or by using extra MCP tools to access the issues. These you can still do but this approach is on purpose different.

Goals

1. Add Github issue support
2. Proof the concept/approach by apply the approach on itself using Github issues for tracking and development of new features
3. Add support for Bugzilla and Redmine using this approach in the process of doing it. Record a video of it.
4. Clean-up and test the implementation and create some documentation
5. Create a blog post about this approach

Resources

There is a prototype implementation here. This currently sort of works with JIRA only.

---

Extended private brain - RAG my own scripts and data into offline LLM AI by tjyrinki_suse

Description

For purely studying purposes, I'd like to find out if I could teach an LLM some of my own accumulated knowledge, to use it as a sort of extended brain.

I might use qwen3-coder or something similar as a starting point.

Everything would be done 100% offline without network available to the container, since I prefer to see when network is needed, and make it so it's never needed (other than initial downloads).

Goals

1. Learn something about RAG, LLM, AI.
2. Find out if everything works offline as intended.
3. As an end result have a new way to access my own existing know-how, but so that I can query the wisdom in them.
4. Be flexible to pivot in any direction, as long as there are new things learned.

Resources

To be found on the fly.

Timeline

Day 1 (of 4)

• Tried out a RAG demo, expanded on feeding it my own data
• Experimented with qwen3-coder to add a persistent chat functionality, and keeping vectors in a pickle file
• Optimizations to keep everything within context window
• Learn and add a bit of PyTest

Day 2

• More experimenting and more data
• Study ChromaDB
• Add a Web UI that works from another computer even though the container sees network is down

Day 3

• The above RAG is working well enough for demonstration purposes.
• Pivot to trying out OpenCode, configuring local Ollama qwen3-coder there, to analyze the RAG demo.
• Figured out how to configure Ollama template to be usable under OpenCode. OpenCode locally is super slow to just running qwen3-coder alone.

Day 4 (final day)

• Battle with OpenCode that was both slow and kept on piling up broken things.
• Call it success as after all the agentic AI was working locally.
• Clean up the mess left behind a bit.

Blog Post

Summarized the findings at blog post.

---

Creating test suite using LLM on existing codebase of a solar router by fcrozat

Description

Two years ago, I evaluated solar routers as part of hackweek24, I've assembled one and it is running almost smoothly.

However, its code quality is not perfect and the codebase doesn't have any testcase (which is tricky, since it is embedded code and rely on getting external data to react).

Before improving the code itself, a testsuite should be created to ensure code additional don't cause regression.

Goals

Create a testsuite, allowing to test solar router code in a virtual environment. Using LLM to help to create this test suite.

If succesful, try to improve the codebase itself by having it reviewed by LLM.

Resources

Solar router github project

---

Bugzilla goes AI - Phase 1 by nwalter

Description

This project, Bugzilla goes AI, aims to boost developer productivity by creating an autonomous AI bug agent during Hackweek. The primary goal is to reduce the time employees spend triaging bugs by integrating Ollama to summarize issues, recommend next steps, and push focused daily reports to a Web Interface.

Goals

To reduce employee time spent on Bugzilla by implementing an AI tool that triages and summarizes bug reports, providing actionable recommendations to the team via Web Interface.

Project Charter

Bugzilla goes AI Phase 1

Description

Project Achievements during Hackweek

In this file you can read about what we achieved during Hackweek.

Project Achievements

---

Self-Scaling LLM Infrastructure Powered by Rancher by ademicev0

Self-Scaling LLM Infrastructure Powered by Rancher

---

Description

The Problem

Running LLMs can get expensive and complex pretty quickly.

Today there are typically two choices:

1. Use cloud APIs like OpenAI or Anthropic. Easy to start with, but costs add up at scale.
2. Self-host everything - set up Kubernetes, figure out GPU scheduling, handle scaling, manage model serving... it's a lot of work.

What if there was a middle ground?

What if infrastructure scaled itself instead of making you scale it?

Can we use existing Rancher capabilities like CAPI, autoscaling, and GitOps to make this simpler instead of building everything from scratch?

Project Repository: github.com/alexander-demicev/llmserverless

---

What This Project Does

A key feature is hybrid deployment: requests can be routed based on complexity or privacy needs. Simple or low-sensitivity queries can use public APIs (like OpenAI), while complex or private requests are handled in-house on local infrastructure. This flexibility allows balancing cost, privacy, and performance - using cloud for routine tasks and on-premises resources for sensitive or demanding workloads.

A complete, self-scaling LLM infrastructure that:

• Scales to zero when idle (no idle costs)
• Scales up automatically when requests come in
• Adds more nodes when needed, removes them when demand drops
• Runs on any infrastructure - laptop, bare metal, or cloud

Think of it as "serverless for LLMs" - focus on building, the infrastructure handles itself.

How It Works

A combination of open source tools working together:

Flow:

• Users interact with OpenWebUI (chat interface)
• Requests go to LiteLLM Gateway
• LiteLLM routes requests to:
  • Ollama (Knative) for local model inference (auto-scales pods)
  • Or cloud APIs for fallback

---

Kubernetes-Based ML Lifecycle Automation by lmiranda

Description

This project aims to build a complete end-to-end Machine Learning pipeline running entirely on Kubernetes, using Go, and containerized ML components.

The pipeline will automate the lifecycle of a machine learning model, including:

• Data ingestion/collection
• Model training as a Kubernetes Job
• Model artifact storage in an S3-compatible registry (e.g. Minio)
• A Go-based deployment controller that automatically deploys new model versions to Kubernetes using Rancher
• A lightweight inference service that loads and serves the latest model
• Monitoring of model performance and service health through Prometheus/Grafana

The outcome is a working prototype of an MLOps workflow that demonstrates how AI workloads can be trained, versioned, deployed, and monitored using the Kubernetes ecosystem.

Goals

By the end of Hack Week, the project should:

1. Produce a fully functional ML pipeline running on Kubernetes with:

   • Data collection job
   • Training job container
   • Storage and versioning of trained models
   • Automated deployment of new model versions
   • Model inference API service
   • Basic monitoring dashboards

2. Showcase a Go-based deployment automation component, which scans the model registry and automatically generates & applies Kubernetes manifests for new model versions.

3. Enable continuous improvement by making the system modular and extensible (e.g., additional models, metrics, autoscaling, or drift detection can be added later).

4. Prepare a short demo explaining the end-to-end process and how new models flow through the system.

Resources

Project Repository

Updates

1. Training pipeline and datasets
2. Inference Service py

---

MCP Server for SCC by digitaltomm

Description

Provide an MCP Server implementation for customers to access data on scc.suse.com via MCP protocol. The core benefit of this MCP interface is that it has direct (read) access to customer data in SCC, so the AI agent gets enhanced knowledge about individual customer data, like subscriptions, orders and registered systems.

Architecture

Goals

We want to demonstrate a proof of concept to connect to the SCC MCP server with any AI agent, for example gemini-cli or codex. Enabling the user to ask questions regarding their SCC inventory.

For this Hackweek, we target that users get proper responses to these example questions:

• Which of my currently active systems are running products that are out of support?
• Do I have ready to use registration codes for SLES?
• What are the latest 5 released patches for SLES 15 SP6? Output as a list with release date, patch name, affected package names and fixed CVEs.
• Which versions of kernel-default are available on SLES 15 SP6?

Technical Notes

Similar to the organization APIs, this can expose to customers data about their subscriptions, orders, systems and products. Authentication should be done by organization credentials, similar to what needs to be provided to RMT/MLM. Customers can connect to the SCC MCP server from their own MCP-compatible client and Large Language Model (LLM), so no third party is involved.

Milestones

[x] Basic MCP API setup
  MCP endpoints
  [x] Products / Repositories
  [x] Subscriptions / Orders 
  [x] Systems
  [x] Packages
[x] Document usage with Gemini CLI, Codex

Resources

• fast_mcp library for Rails
• MCP test client
• SCC branch

Gemini CLI setup:

~/.gemini/settings.json:

---