Gökdeniz Gülmez
Portrait of Gökdeniz Gülmez
Open source · Apple Silicon

Open-source contributor · Apple MLX

Gökdeniz
Gülmez

Systems Engineer at Computacenter, specialising in machine learning engineering and MLOps. I build machine-learning systems and help move them from experiments into reliable services. Outside work, I contribute to open source and research ways to run and train AI on Apple chips. I’m a top MLX-LM contributor and, as a freelancer officially sponsored by Neywa Labs, the sole maintainer of MLX-VLM’s core trainer. I also build the J.O.S.I.E. model family and JOSIE-Voice for Mac.

Apple chips Apple’s MLX tools Training language models MLOps Local AI

I help run and train AI
on Apple chips.

At Computacenter, I work as a System Engineer focused on machine-learning engineering and MLOps—the work of deploying, monitoring and maintaining software that uses AI. Outside work, I contribute to open source and conduct ML research. Apple’s MLX framework is a foundation for building, training and running machine-learning models on Apple chips. MLX-LM adds text language models; MLX-VLM adds models that work with images and words and is now developed at Neywa Labs, which officially sponsors my freelance work on its core trainer. My contributions help those libraries support more models and training methods.

01 / SystemsML engineering & MLOps
02 / RuntimeRun models on Apple devices
03 / TrainingAdapt models to new tasks
04 / ResearchHow models work and specialize

02 / Selected impact

Built for scale.
Proven in real use.

Top 5

MLX-LM contributor

Around 10,000 lines of code contributed to Apple’s open-source language model toolkit.

25+

Model variants supported

MLX-LM architecture work spanning more than 15 organizations.

30K

Lines added to the MLX-VLM trainer

As a freelance engineer officially sponsored by Neywa Labs, I’m the sole maintainer of the core trainer, with SFT, DPO, ORPO and GRPO workflows. MTP and QAT are in open pull requests as the trainer expands to more modalities.

100+

Tool calls in one JOSIE-Voice demo

A six-minute MacBook Pro M4 Pro call, shown in the teaser below, with on-device speech models and a cloud reasoning model.

Watch the teaser ↘

03 / Upstream

I make AI work
on Apple chips.

MLX is Apple’s open-source toolkit for machine learning. The tools around it help people run and train AI models directly on Apple chips. Here are some of my credited contributions.

01Core framework · MLX

New training primitives

I added Muon, a method for updating a model as it learns, and ReLU², a small mathematical function used inside neural networks. Both are credited in MLX’s contributor acknowledgements.

See MLX acknowledgements
02Language models · MLX-LM

Support for 20+ model designs

MLX-LM is an open-source library in Apple’s MLX ecosystem for running and fine-tuning text-based language models on Apple Silicon. I’m one of its top five contributors, with around 10,000 lines of code added in total. I helped it support more than 20 model designs, and added ways to train every learned value in a model, choose between training methods, and track experiments with Weights & Biases, a tool for recording and comparing training runs. A model design is its internal blueprint; the list below links to the original model files and code changes.

Read MLX-LM credits
03Vision-language · MLX-VLM · Neywa Labs

Core trainer for MLX-VLM at Neywa Labs

MLX-VLM extends the MLX ecosystem to vision-language models that understand images and text together. It now belongs to Neywa Labs, the company building local AI infrastructure for Apple Silicon. As a freelance engineer officially sponsored by Neywa Labs, I develop, create, set up and maintain MLX-VLM’s core trainer, and have added 30,000 lines of code. The trainer supports SFT, DPO, ORPO and GRPO; MTP and QAT are in open pull requests, with more modalities and algorithms planned.

Explore MLX-VLM Visit Neywa Labs See selected trainer PRs
04Examples · MLX Examples

Workflows people can run

MLX Examples offers practical, runnable demonstrations. Its acknowledgements credit my work bringing several models to the examples and adding full training, which adjusts all of a model’s learned values.

Read MLX Examples credits
Model architecture support · MLX-LM

20+ model variants across 15+ organizations.

MLX-LM credits my work on these 25+ model variants from 15+ organizations. An architecture is a model’s internal design; weights are the learned values that let it work. “Weights” links to original model files. “PR” means pull request: a proposed code change and its review.

Qwen3 note: PR #42 closed before it was merged. The project credits my help on Qwen3 and its version built from specialist submodels; PR #199 is a later accepted update to that version.

Read the MLX-LM acknowledgements

04 / Research

Researching how
models learn and act.

Preprint · 2026

DynaMoE

A model with several specialist parts that can activate different experts for different inputs. DynaMoE studies whether choosing those specialists dynamically can help a model use its computing resources more efficiently.

Read the paper
Preprint · 2025

Gabliteration

A method for changing selected model behaviors by editing its learned values, while checking that other abilities still work well.

Read the paper
AI training method · ongoing

Learning from preferred answers

Directional and Similarity-aware Latent Alignment (DSLA) studies how feedback about better answers can also shape the patterns a model forms internally while processing text.

Read the paper

05 / Model family

J.O.S.I.E. —
models with character.

My open model family · Hugging Face

Reasoning-first models, built locally.

J.O.S.I.E. is my family of downloadable AI models, exploring reasoning, honesty and independent judgment. J.O.S.I.E.-2 includes models with 2, 4 and 9 billion learned values, trained on Apple chips. Smaller versions, with fewer bits used to store each learned value, need less memory and are easier to run on a personal computer.

Read the J.O.S.I.E.-2 research note Visit the J.O.S.I.E. project page

06 / JOSIE

Talk to your Mac.
Let it take action.

In development · macOS · Apple Silicon

A local, real-time, full-duplex voice and chat assistant that listens, knows who's speaking, talks back with a natural voice, and takes real action in your Reminders, Calendar, Maps and Messages.

JOSIE-Voice listens and speaks on your Mac, remembers conversations, and can act in Reminders, Calendar, Maps and Messages. The teaser shows a six-minute call on an M4 Pro with more than 100 tool calls: Qwen-3 0.6B handles on-device speech recognition and text-to-speech, while DeepSeek 4.1 Flash runs through Ollama Cloud.

Teaser · October 18, 2026

07 / Selected work

Tools, models
& experiments.

Vision-language training · Neywa Labs

MLX-VLM

MLX-VLM runs and trains models that work with images and text on Apple Silicon, and is now part of Neywa Labs. As a freelancer officially sponsored by Neywa Labs, I develop, set up and maintain its core trainer, with SFT, DPO, ORPO and GRPO workflows. MTP and QAT are in open pull requests as the trainer expands to more modalities.

Explore the project Neywa Labs
Native Mac app

MLX-LoRA-Studio

An open-source Mac app for adapting language models to new tasks, checking their results and managing models on Apple chips.

Explore the project
Model research

J.O.S.I.E.

A personal model family exploring small language models, reasoning and assistant behavior, with releases and research notes on Hugging Face.

Explore J.O.S.I.E.
Mac voice assistant

JOSIE-Voice

An in-development Mac voice and chat assistant with on-device speech recognition and speech, conversation memory, and actions in everyday apps.

See the teaser Explore the project
Alignment research

DSLA

Directional and Similarity-aware Latent Alignment studies how preference feedback can shape a model’s internal representations as well as its answers.

Read the paper draft Explore DSLA
Interpretability

MLX-LM-LENS

A research tool for looking inside language models: how they represent words, which parts of the input they focus on, and how information moves through them.

Explore the project
Local AI

Local NotebookLM

A notebook that uses your PDFs as the source for summaries and creates audio versions with language and speech tools you choose.

Explore the project
Model editing

Gabliteration

Research and tools for finding model edits that change selected behaviors while aiming to keep other abilities intact.

Explore the project
MLX research

MLX-KAN

An implementation of Kolmogorov–Arnold Networks, an alternative model design built from simple, learnable functions, using Apple’s MLX toolkit.

Explore the project

08 / Career

Career

Computacenter

System Engineer

System Engineer focused on building machine-learning software and keeping it reliable after deployment. MLOps covers the tools and practices for shipping, monitoring and maintaining AI systems.

In my free time, I contribute to open source and pursue ML research and engineering.

09 / References

Upstream credits,
papers & releases.

10 / Elsewhere

Follow the work
where it lives.

I post updates, new research and project releases on X at @ActuallyIsaak.