MacPaw is a software company that develops and distributes software for macOS and iOS. Today, we have 20 million active users across all our products.
At MacPaw, we believe humans and technology can reach their greatest potential together.
MacPaw is proud to be Ukrainian. The support and development of Ukraine are significant parts of the company’s culture. MacPaw gathers open-minded people who support each other and aspire to change the world around us.
Imagine working on an AI-powered macOS assistant that feels truly intuitive, smart, and seamlessly integrated into users' daily workflows. At MacPaw, we’re pushing LLM-powered agents to the next level, and we need a Senior Data Scientist to help us get there.
In this role, you won’t just build models—you’ll shape the next generation of on-device AI, tuning LLMs to achieve SOTA performance, RAG pipelines, and intelligent agents for real-world applications. You’ll work with top Engineers and Scientists to bring cutting-edge NLP and ML capabilities directly to macOS users.
If you’re passionate about LLMs, AI agents, and making AI more accessible and powerful, this is your chance to be at the forefront of a game-changing innovation.
P.S. Here, you can learn more about our macOS AI assistant.
In this role, you will:
- Fine-tune LLMs
- Develop and fine-tune LLM-based agents for local task execution
- Design and build advanced RAG (Retrieval-Augmented Generation) pipelines for improved information retrieval
- Lead general deep-learning training and experimental initiatives. Cooperate with and mentor students from top Ukrainian universities
- Apply classical NLP techniques (e.g., NER, POS tagging)
- Set up a testing framework to evaluate agent performance, measure key metrics, and identify optimization opportunities
- Supervise macOS Engineers to integrate models on device
Skills you’ll need to bring:
- Advanced experience in NLP and LLMs
- Proven ability to train custom NLP models and fine-tune pre-trained LLMs
- Strong background in Transformer architectures and their applications
- Proficiency in classical ML tasks (classification, regression, clustering)
- Experience with Python and NLP/ML frameworks (PyTorch, spaCy, NLTK, SciPy)
- Hands-on experience building LLM agents (RAG, ReAct, Plan&Execute, multi-agent systems)
- Good expertise in creating competitive advantage through LLM fine-tuning and RAG optimization
- Familiarity with LLM agent frameworks (LangChain, LangGraph, Vector Databases)
- Experience implementing metrics frameworks that align AI performance with business outcomes
- At least an Intermediate level of English
As a Plus:
- Knowledge of running and optimizing open-source LLMs
- Background in conversational AI and dialog systems
- Familiarity with iOS/macOS ML frameworks for local deployment
- Experience with edge deployment of ML models
- Expertise with LangFuse or similar observability tools
