Run your LLM locally and turn them into Agents

Onuralp SEZER
I'm Onuralp, a seasoned Senior Software Engineer Team Lead with expertise in Python, Kotlin, C++, and Rust. My passion lies in the dynamic fields of Computer Vision, Machine Learning, Deep Learning. I am a long term Fedora Contributor,Fedora KDE SIG Member and others roles at Fedora Project. I am also an organizer at GDG Samsun and speaker related around Machine learning and Computer Vision. Professionally, I use Python development on backend side of works and on mobile side I used Kotlin and Flutter with Google technologies. ML sides I used Tensorflow, Tensorboard, Keras, Jax, Pytorch. In addition to my diverse skill set, I am actively contributing to a computer vision project named Supervision. Supervision is dedicated to crafting reusable computer vision tools tailored to a wide array of needs. This project embodies my commitment to advancing the field of computer vision, offering robust solutions that empower developers and researchers alike.

Onur Can KARAMAN
Onur Can Karaman is an industrial applications leader with a strong focus on digital transformation, artificial intelligence (AI), and the Internet of Things (IoT). Currently at Digitheta, he leads digitalization projects for manufacturing enterprises, driving innovation in operational efficiency and data-driven decision-making. Previously, he served as the Digital Technologies Specialist. He played a key role in promoting digital capability development across the industrial sector. With a solid background in lean manufacturing.
As a certified digital transformation consultant trained through a national-level program, Onur has experience in assessing digital maturity and designing strategic roadmaps for industrial companies. His technical interests lie in deploying large language models (LLMs), computer vision, and AI-powered solutions such as predictive maintenance, anomaly detection, and digital quality control.
He is also a contributor to open-source projects, including OpenCV, reflecting his commitment to collaborative innovation and continuous learning.
No video of the event yet, sorry!
Large Language Models (LLMs) are powerful tools, but running them locally unlocks new possibilities—enhancing privacy, reducing costs, and allowing full customization. In this talk, we’ll explore how to set up and optimize LLMs on your machine, leveraging open-source models and efficient inference techniques. We’ll then take it a step further by turning these models into intelligent agents that can interact with external tools, retrieve information, and automate tasks. Whether you're a developer, researcher, or AI enthusiast, this session will provide practical insights into running LLMs locally and building AI-driven agents that work for you.
- Date:
- 2025 June 26 - 16:00
- Duration:
- 1 h
- Room:
- Seminar Room 1
- Conference:
- openSUSE Conference 2025
- Language:
- Track:
- Open Source
- Difficulty:
- Medium
- Part 4: CRA/NIS2 Readiness for Open Source Projects and SME Vendors
- Start Time:
- 2025 June 26 15:15
- Room:
- Seminar Room 2
- Fine tuning log routing
- Start Time:
- 2025 June 26 16:00
- Room:
- Gallerie
- The Great Migration? (Part 1)
- Start Time:
- 2025 June 26 16:00
- Room:
- Saal
- Swift Deployment and Faster Releases
- Start Time:
- 2025 June 26 16:45
- Room:
- Saal
- Bridging the Digital Divide
- Start Time:
- 2025 June 26 16:45
- Room:
- Gallerie