Working Student AI/ML

About Tensordyne (formerly Recogni)

Artificial intelligence (AI) is transforming our world. It can perform cognitive functions that previously only humans could do, such as perceiving interactions across different modalities and environments - with the ability to quickly learn and then solve complex problems. Tensordyne is an AI system solution company that builds very high-performance, low-power generative AI inference systems. Our mission, through the creation of custom silicon, hardware and software, is to enable multimodal Generative AI inference acceleration at scale, with safe, sustainable, high-performance systems for our hyperscaler and neocloud data center customers. We are at the leading edge of advancing the latest research and product improvements for generative Al inference solutions that will make Al even more advantageous for compelling new generative AI applications.Tensordyne is a well funded, fast-paced startup company with headquarters in both Sunnyvale, CA, and Munich, Germany. We also have many talented team members working remotely across North America and Europe.

The Opportunity

At Tensordyne, you'll have the opportunity to work at the forefront of generative AI, engaging with large language models (LLMs) as well as multimodal systems for video, image, and speech. You'll contribute directly to developing features that deliver blazing-fast, high-accuracy inference on our custom-built hardware.

As a working student, you'll gain valuable experience in a dynamic and innovative environment that thrives on curiosity and collaboration. This part-time role (20 hours per week) is based on-site at our Munich office. We encourage knowledge sharing, with opportunities to publish your work as a conference paper or blog post to enhance its impact.

The specific duties of this role will be shaped by both our business priorities and your individual interests.

What you'll bring

• Strong understanding of generative AI models such as LLMs and Diffusion models, with the ability to evaluate and quantify their performance.
• Familiarity with model quantization techniques, using both open-source tools and proprietary approaches.
• Ability to analyze and interpret model architectures, identifying nuances and edge cases in performance profiling.
• Interest in multi-modal AI and motivation to extend our model zoo with the SOTA models.
• Clear communication skills for documenting results and collaborating effectively with the AI team and across the company.
• Curiosity and initiative to stay current in the field by engaging with research papers, technical blogs, and community events.

Qualifications

• Currently enrolled in a Master's program in Electrical Engineering, Robotics, Computer Science, or a related AI-focused field.
• Solid understanding of modern deep learning models and methodologies. Knowledge of DNN hardware acceleration and model compression techniques such as low-precision quantization or profiling is a strong plus.
• Practical experience developing AI applications through internships, academic projects, personal work, or other engagements. Prior exposure to generative AI is preferred.
• Strong ability to analyze and distill insights from research publications in the AI field.
• Proficiency with core tools and frameworks, ideally including:
• PyTorch
• Python
• Git
• Rust
• Ability to thrive in a fast-paced, multidisciplinary environment. Successful candidates are curious, embrace challenges, and are committed to continuous learning and improvement.