Data scientist/chemist (H/F)

SATT AxLR is based in the Mediterranean arc of Occitanie Region (France). Its mission is to increase the intensity of technology transfer from public research while improving innovation processes.

Context of the project:

The integration of artificial intelligence in chemical synthesis processes is a key lever to transform how reactions are designed, optimized and industrialized. Through machine learning and Bayesian optimization, experimental parameters can be correlated in real time with process performance to rapidly identify optimal operating conditions. This significantly reduces the number of experiments, reagent consumption and experimental time, while improving reproducibility and result quality. More broadly, AI contributes to a more sustainable and agile approach to chemistry, enhancing R&D responsiveness, reducing development costs and enabling intelligent use of historical data.

Mission :

Design, develop, and deploy a Bayesian optimization system for a chemical synthetic process. More specifically, the objective is to develop an optimization framework for mechanochemical processes—methods of chemical synthesis using milling equipment (ball mills, extrusion, etc.). The system should enable real-time monitoring of experiments by Process Analytical Tools (PAT) and automatically identify optimal operating conditions based on experimental data.

The assignment will also involve automating certain steps to reduce manual effort to the strict minimum.

The project requires an interdisciplinary profile combining preferably, expertise in mechanochemistry, or alternatively in chemistry, real-time analytical techniques, experimental design, data analysis and programming.

The selected candidate will work in the interface between experimental chemistry and AI tools to develop autonomous optimization. He/she will be responsible for reporting to both the SATT project manager and the scientific lead of the project, providing all necessary information to ensure effective project monitoring and execution (experimental organization, progress reports, project delays, expenditures, etc.).

Main activities :

• Integrate and automate the collection of data from synthesis equipment (temperature, weighing, yield, characterization, etc.)

• Establish a communication interface between experimental instruments and the computational platform

• Analyze and visualize results to guide future experimental work

• Develop and implement in-situ monitoring techniques

• Work closely with experimental/process chemists to define relevant parameters and interpret outcomes

• Monitor project progress within the laboratory, with regular reporting to the SATT AxLR project manager and the scientific lead, in order to optimize project resource allocation
  
  

Associated activities :

• Prepare presentations and any other supporting materials for meetings with partners, prospects, and investors

• Present the project’s technical results to all stakeholders during progress review meetings

• Contribute to the drafting of scientific and technical reports to promote and disseminate results

• Write progress reports (interim and final reports)
  
  

Main skills:

• Proficiency in programming languages such as C++, Python, and MATLAB

• Experience with Bayesian optimization libraries ( e.g., GPy, BoTorch, GPflow, PyMC3)

• Strong understanding of the principles of Bayesian optimization and machine learning

• Knowledge of experimental data acquisition and management systems (APIs, databases, communication protocols) applied to chemistry

• Solid skills in data analysis, data processing, modelling and interactive visualization (Plotly, Dash, Streamlit)

• Ability to work at the interface between data science and experimental chemistry

• A background in chemistry is a significant advantage for this position (e.g., knowledge of organic chemistry, reaction mechanisms, interpretation of experimental data by solution-based and solid-state characterization techniques, including PXRD, in-situ monitoring of mechanochemical processes by RAMAN spectroscopy).

• Knowledge of mechanochemistry is a plus
  
  

Additional skills :

• Basic knowledge of process chemistry and an understanding of experimental constraints would be highly valued

• Experience with version control tools and collaborative development environments (e.g., GitHub)

• Ability to document code and adhere to software development best practices (e.g., maintaining a project wiki)

• Ability to work effectively in an interdisciplinary and collaborative environment (international team)

• Proficiency in written and spoken English is a must.

• Proactive mindset and ability to take initiative
  
  

Degree :

PhD in process engineering, chemistry, data science, or a related field, with applications in optimization (particularly Bayesian optimization), machine learning, or process control. Candidates with additional expertise in programming, data science, real-time monitoring, machine learning or process automation are strongly encouraged to apply.

Experience :

• 2–3 years of experience (extended internships, postdoctoral fellowship, or first professional position) in modeling, programming, optimization, or data science applied to physico-chemical systems or industrial processes, synthetic chemistry and real-time in situ monitoring technics

• Proven experience in experimental optimization (Design of Experiments—DoE, sequential optimization, ideally Bayesian optimization) in a laboratory or pilot-scale environment

• Proven experience in in-situ monitoring techniques, elucidation of reaction mechanisms, kinetic modelling is a plus

• Involvement in laboratory and/or pilot scale Active Pharmaceutical Ingredients (API) development would be highly appreciated

• Experience working directly with experimental R&D teams (chemists, process engineers)

• Experience in data acquisition/management systems (databases, instrument APIs, ELNs, etc.)
  
  

Interviews will be conducted with SATT AxLR project manager and project team members