Maximilian Winter
Unity Developer at DigitalTwin Technology GmbH
Königswinter, North Rhine-Westphalia, Germany
Summary
👤
Senior🎓
Top SchoolMaximilian Winter is a Unity developer with eight years of professional experience building 2D, 3D, VR, AR and mixed-reality applications, currently contributing at DigitalTwin Technology after several years at Nurogames. He began developing interactive graphics at 14 and has a strong systems background in C++, C#, DirectX and Unity, with hands-on experience porting CAD visualization to HoloLens during an ABB proof-of-concept. Beyond games and immersive apps, he actively explores AI and deep learning, contributing to llama-cpp-python by improving sampling, logit processing and cuBLAS-based performance optimizations for LLM inference. He combines low-level performance tuning with practical product delivery—designing, implementing, testing and supporting production VR systems. Based in Königswinter, Germany, he brings a curious, research-minded approach to applied engineering and tool-building across graphics and ML.
9 years of coding experience
5 years of employment as a software developer
Fachinformatiker für Anwendungsentwicklung, Fachinformatiker für Anwendungsentwicklung at AFP
Github Skills (11)
machine-learning10
api10
python10
apidoc10
llm-inference10
numpy9
cublas9
imp8
im8
docker4
dockers4
Programming languages (8)
Github contributions (5)
abetlen/llama-cpp-python
May 2023 - Feb 2025
Python bindings for llama.cpp
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:4 reviews, 7 PRs, 10 comments in 1 year 9 months
Contributions summary:Maximilian primarily focused on enhancing the `llama-cpp-python` project by adding and refining features related to model sampling and inference. This involved implementing mirostat sampling methods within the high-level API and integrating new functionalities for logit processors and stopping criteria. Furthermore, they introduced improvements, such as enabling cuBLAS, which optimizes matrix operations. Their contributions demonstrate expertise in modifying the core logic for LLM interaction and supporting the project's performance.
python-bindings
Maximilian-Winter/llama-cpp-agent
Dec 2023 - Mar 2026
The llama-cpp-agent framework is a tool designed for easy interaction with Large Language Models (LLMs). Allowing users to chat with LLM models, execute structured function calls and get structured output. Works also with models not fine-tuned to JSON output and function calls.
Contributions:72 releases, 1 review, 27 PRs in 2 years 2 months
agent-frameworkfunction-calllarge-language-modelsllamacppllm
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.