Student AI Engineer – Finetuning LLMs (20 hours/week)

Stellenticket GmbH; Berlin; Hybrid; EUR 40.000 - 80.000; Vollzeit

Zusammenfassung

Ein innovatives Startup sucht leidenschaftliche Talente mit einem soliden KI-Hintergrund, um in einem dynamischen Umfeld zu wachsen. Diese Rolle bietet die Möglichkeit, an der Entwicklung von KI-Technologien zu arbeiten und Erfahrungen in der Feinabstimmung von großen Sprachmodellen zu sammeln. Sie werden eng mit einem erfahrenen Senior AI Engineer zusammenarbeiten und haben die Chance, in eine Vollzeitstelle überzugehen. Flexible Arbeitszeiten und die Möglichkeit, remote zu arbeiten, machen diese Position besonders attraktiv. Werden Sie Teil eines Teams, das die Art und Weise revolutioniert, wie Präsentationen erstellt werden, und gestalten Sie die Zukunft der KI mit.

Leistungen

Flexible Arbeitszeiten

Remote-Arbeit

Faire Vergütung

Talentförderungsprogramm

Qualifikationen

• Praktische Erfahrung in der Feinabstimmung von LLMs und Verständnis für PyTorch.
• Aktuell im Bachelor- oder Masterstudium in Informatik oder Data Science.

Aufgaben

• Unterstützen Sie die Entwicklung von KI-Agenten und optimieren Sie offene Modelle.
• Implementieren und verbessern Sie Retrieval-Augmented Generation Systeme.

Kenntnisse

Hugging Face Ecosystem

PyTorch

Fine-tuning LLMs

Multi-modal Models

Retrieval-Augmented Generation (RAG)

Quantization Techniques

Ausbildung

Bachelor in Computer Science

Master in Data Science

Tools

Transformers

Datasets

Jobbeschreibung

The Role:

This position is ideal for ambitious individuals with a solid AI background who are eager to gain hands-on experience in fine-tuning Large Language Models (LLMs). You will work closely with an experienced Senior AI Engineer, leverage cutting-edge technologies, and grow within an innovative startup environment. For candidates nearing graduation, there is also the opportunity to transition into a full-time role.

About Us:

Zapdeck is revolutionizing the way presentations are created. With our AI-powered software, users can generate professionally designed slides in just minutes. Our innovative AI technology transforms text directly into visually compelling slide designs, supported by an intuitive editor that simplifies storyline development and layout design.

We operate in a multi-billion-dollar market that is being redefined by Generative AI. Backed by a pre-seed investment, we are building partnerships with leading companies and working on a multi-modal, agent-based system that utilizes the latest AI models.

Your Responsibilities:

• Support the development of AI agents
• Execute fine-tuning processes for LLMs using an existing server setup and prepared datasets
• Systematically test and optimize open-source models and their hyperparameters
• Implement and improve Retrieval-Augmented Generation (RAG) systems
• Contribute to the development of multi-modal model approaches (text + image)
• Apply modern quantization and fine-tuning techniques (LoRA, QLoRA, Bitsandbytes)

Your Qualifications:

• Currently pursuing a Bachelor's or Master’s degree in Computer Science, Data Science, or a related field
• At least 1 year of practical experience (e.g. from student projects, internships, or relevant work experience)
• Solid understanding of the Hugging Face ecosystem (Transformers, Datasets) and PyTorch
• Hands-on experience in fine-tuning LLMs
• Experience with multi-modal models, RAG, or agent-based systems is a plus
• Ability to work independently and in a structured manner

What We Offer:

• Flexible work arrangements: remote work or one in-person office day per week in Berlin-Mitte, with flexible hours
• Fair compensation and the opportunity to gain valuable real-world experience in a dynamic team
• The chance to contribute to the development of innovative AI technologies and gain insights into building a successful startup
• Clear path to a future full-time role, supported by a talent funding program to partially cover employment costs

We look forward to your application!