Software Developer

Massachusetts Institute of Technology; Cambridge (MA); On-site; USD 110,000 - 170,000; Full time

Job summary

Massachusetts Institute of Technology in Cambridge invites applications for a Chemical Engineering (ChemE) - Machine Learning for Pharmaceutical Discovery and Synthesis (MLPDS) Consortium role. You will help translate scientific workflows into robust, scalable computational applications and work with faculty, researchers, and graduate students.

Responsibilities include professionalizing software developed by team members, maintaining a modern web app for long-running ML tasks, defining API

Responsibilities

• professionalization of software developed by graduate students/postdocs/MLPDS team members
• maintaining and continuing the development of a modern web application for submitting and processing long-running ML tasks and storing, retrieving, visualizing, and analyzing results
• developing API standards and data structures for ML predictions for synthesis planning, property prediction, and molecular design
• defining and codifying reproducible and transferable ELT pipelines for training ML models on chemical data
• preparing scripts for facilitating the automatic retraining and deployment of ML models on user-provided chemical and reaction datasets
• packagingthe application into containerized microservices for deployment using Docker and Kubernetes/EKS
• and monitoring application usage and setting resource limits.

Job description

Chemical Engineering (ChemE)- Machine Learning for Pharmaceutical Discovery and Synthesis (MLPDS) Consortium, to be partially responsible for the continued development and maintenance of command-line and web-based applications deploying machine learning (ML) models for chemical synthesis planning, property prediction, and molecular design and will work closely with faculty, researchers, and graduate students to translate scientific and engineering workflows into robust, scalable, and reproducible computational applications.

Responsibilities

• professionalization of software developed by graduate students/postdocs/MLPDS team members
• maintaining and continuing the development of a modern web application for submitting and processing long-running ML tasks and storing, retrieving, visualizing, and analyzing results
• developing API standards and data structures for ML predictions for synthesis planning, property prediction, and molecular design
• defining and codifying reproducible and transferable ELT pipelines for training ML models on chemical data
• preparing scripts for facilitating the automatic retraining and deployment of ML models on user-provided chemical and reaction datasets
• packagingthe application into containerized microservices for deployment using Docker and Kubernetes/EKS
• and monitoring application usage and setting resource limits.