IMC
Deep Learning Research Intern - Summer 2027 - Amsterdam
Amsterdam, Netherlands +2 · Internship · Posted Jul 6, 2026
- Location
- Amsterdam, Netherlands
New York, United States
Sydney, Australia - Education
- Current PhD or Master’s student in Computer Science, Machine Learning, Artificial Intelligence, Mathematics, Statistics, Engineering, or a related quantitative field; graduating in 2028.
- Visa sponsorship
- IMC cannot obtain immigration sponsorship for candidates who hold citizenship from Russia, Belarus, or Iran and attend university in one of those countries. Candidates with Ukrainian citizenship must confirm they hold a biometric passport for sponsorship. Candidates residing or attending university elsewhere are encouraged to apply if they already have Dutch work authorization or can work without sponsorship.
About the job
Conduct deep learning research with IMC researchers, building or improving models with potential business impact using large-scale datasets and modern training infrastructure.
What you'll do
- Collaborate with Deep Learning Researchers on projects to build or improve models.
- Explore research areas including representation learning, transformers, foundation models, generative AI, and reinforcement learning.
- Develop research skills with support from a mentor and intern lead.
What they ask for
- Currently pursuing a PhD or Master’s degree in a listed or related quantitative field and graduating in 2028.
- Experience applying deep learning techniques.
- Strong Python programming skills and hands-on experience with at least one modern deep learning framework.
- Able to start in person during the last week of June 2027.
Nice to have
- Publications at leading ML venues such as NeurIPS, ICML, or ICLR.
- Contributions to open-source research.
Tools
Python, PyTorch, JAX, TensorFlow
AI focus
Deep learning research involving model development and improvement, including representation learning, transformers, foundation models, generative AI, and reinforcement learning.
Model-written summary · the original posting is authoritative