Drug Discovery Optimization & MLOps Platform for Global Top 20 Pharmaceutical Firm

Published: April 10, 2024

Learn how Reveal's MLOps platform helped a leading pharmaceutical company train, deploy, run, manage, and monitor hundreds of machine learning (ML) models to optimize drug discovery efforts.

Implementing Multiple ML Models at Scale with Existing Infrastructure

Our client wanted to leverage ML models to aid in drug discovery. However, ML models need a large number of labeled datasets for training. Additionally, deploying these ML models on the client’s available infrastructure represented changes in scalability & bandwidth.

Reveal’s MLOps Solution

After collaborating with our client to review these challenges in detail, Reveal developed a customized MLOps platform for their needs. This platform leveraged off-the-shelf components such as SageMaker, MLFlow, and Grafana with appropriate tweaks to support our client’s unique circumstances. It was able to train, deploy, run, manage, and monitor hundreds of ML models at scale. The platform was built with APIs that had appropriate access controls to support the needs of different teams throughout the organization, both technical & non-technical.

Positive Client Results

Once the MLOps platform was built and implemented, the client observed the following improvements:

1. Access control became simple to implement, thereby helping improve efficiency across teams
2. Training, deploying, running, managing and monitoring hundreds of ML models became feasible within the client’s infrastructure at a nominal investment
3. The client was able to focus on their most important priority – drug discovery – with much faster turnaround cycles

‍