Iktos is an applied artificial intelligence startup that helps chemistry researchers improve their molecular discovery, drug design process and synthetic planning.
Our technology leverages many advances in machine learning to generate novel molecules that satisfy the end user’s criteria in a fraction of the time and effort. A second area of research is related to data-driven retrosynthetic analysis and empowers organic chemists with exhaustive analyses in minutes.
To support Iktos’ fast-growing team and keep pace with our high number of cutting-edge projects, we are recruiting passionate AI interns. If you wish to join a young team, experience the startup mindset and be mentored by top-notch scientists while working on exciting projects, reach out to us !
Our R&D team is focused on various subjects to directly improve what we offer to our customers, depending on your interest and skills you might work on:
- Studying and experimenting various methods to improve our generative models
- Automated Chemical Reaction Extraction from patents : named entity recognition over partially labeled patents texts
- Implementing and training deep learning generative models using 3D and/or force field information
- Build and train binding affinity prediction models between proteins-ligands pairs. Ensure the generalization power of the model
- Improve one of our current generative models with an auto-encoder architecture: exploration and optimization in the latent space
We are looking for students in their final year of a MSc in data science or Engineering degree. Candidates may have a computer science/software engineering or a probability/statistics/data science background, provided that they are motivated to expand their skill set.
- Knowledge in supervised/unsupervised learning methods
- Advanced level in Python language and being familiar with a deep learning framework (pytorch, tensorflow …)
- Knowledge in deep learning and optimization
- Understanding of basic concepts in chemistry is a plus
- Ease with communication/explanation of your work and results (both in French and in English)
De novo molecular design and generative models. J. Meyers, B. Fabian, N. Brown (2021)
Planning chemical syntheses with deep neural networks and symbolic AI. Nature. Segler, M., Preuss, M. & Waller, M. (2018)
- Duration: 6 months (4 months minimum)
- Location: Paris (17th district) and partial remote
Please send your CV, cover letter and availability by following the link below: