Location: Spain
Salary: €65k-€80k base + bonuses + relocation package
About the Company:
A rapidly growing deep-tech firm creating advanced AI solutions that enhance efficiency and performance across sectors such as finance, energy, and manufacturing. Our diverse, international team thrives on collaboration and innovation. Role Overview:
We are looking for an LLM Engineer to develop, refine, and deploy large language models. You will work on state-of-the-art AI projects, integrating models into products and solving complex, practical challenges.
Key Responsibilities:
- Design and implement techniques to make LLMs more efficient, scalable, and effective.
- Conduct thorough evaluations and optimizations to improve model accuracy, speed, and reliability.
- Identify opportunities for innovation and propose creative solutions.
- Maintain clear documentation of experiments and outcomes; support and mentor junior engineers.
- Keep up to date with advances in LLM research and emerging AI technologies.
- Master’s or Ph.D. in AI, Computer Science, Data Science, or related field.
- Mid: 2+ years hands-on experience with transformer models; Senior: 5+ years including other deep learning architectures.
- Strong Python programming skills and experience with PyTorch, HuggingFace, and cloud/deployment tools.
- Solid understanding of deep learning concepts, model training, and GPU performance optimization.
- Excellent communication skills; fluent in English.
- Experience delivering AI solutions in industry.
- Spanish language skills and familiarity with large-scale cloud/HPC environments.
- Experience evaluating LLMs, building RAG systems, or working on other AI applications (CV, audio, signal processing).
- Awareness of AI ethics and best practices; DevOps/MLOps experience a plus.
- Permanent contract with performance bonus and signing bonus
- Private health coverage, relocation and visa support if needed
- Flexible hours, hybrid work, educational budget, language classes
- Career growth opportunities and collaborative work culture
