Platform Developer I

At Cadence, we hire and develop leaders and innovators who want to make an impact on the world of technology.

OpenEye, Cadence Molecular Sciences, an industry leader in computational molecular design, uses a scientific approach to deliver rapid, robust, and scalable software, toolkits, and technology & design services for the advancement of pharmaceuticals, biologics, agrochemicals, and flavors and fragrances. We are passionate about science and about taking care of our team members, with leaders supporting your development and recognizing your contributions by providing an unmatched total compensation package in a collaborative, caring, energetic and fun work environment.

Over the last eight years, OpenEye has developed the field's most advanced large-scale compute platform, Orion, built on Amazon Web Services. In this position, you will be working at the cutting edge of cloud computing to provide a platform for large-scale investigations such as advanced MD simulation techniques to investigate the mechanistic details of passive membrane permeability of drug-like molecules. Your work will have the potential to make a major impact in drug discovery research and will likely be used by many pharmaceutical and biotech companies around the world.

Orion is a container-based scientific workflow system written in Go and Python. While there are some interesting workflow and container orchestration systems out there, none deliver the flexibility in using legacy applications and toolkits, ease of use, scalability, and reliability that we are targeting. Our system is made up of Floe, a Python workflow framework, and a cloud-based workflow runtime and user interface leveraging many technologies, including Django, Docker, and AWS.

More about Orion and the science being done on it:

https://www.youtube.com/watch?v=gjKNHK4jWMQ

Below is a summary of a few of the scientific problems being solved with Orion:

Molecular Dynamics

• Developed with Industry & Academic s/w, collaboration (OpenMM, Open Force Field)
• Uses a combination of CPUs, GPUs (several types)
• Relies on Elastic parallelism

Crystal Structure Prediction

• Leverages open-source quantum mechanical code: Psi4
• Requires iterating on science & tech. problems
• Each iteration requires 10,000-100,000 CPU hours
• Multiple iterations per day

The largest molecular Docking calculations ever run

• https://docs.eyesopen.com/toolkits/python/dockingtk/docking.html
• 43B molecules (Enamine Real, Stereo Expanded)
• ~24 hours on 30,000 CPUs
• 41TB result

Large Scale Virtual Screening (FastROCS)

• Powered by OpenEye Toolkits: https://docs.eyesopen.com/toolkits/python/fastrocstk/introduction.html
• Searches 10^11 conformers
• Less than one-hour search, including setup and teardown of machines
• 43B molecules (Enamine Real) searched routinely in about 30 minutes

What you should have:

• B.S. in Computer Science or equivalent experience in addition to
• Practical experience with algorithms used in Distributed Systems
• Expertise working with Amazon Web Services and remote diagnostics
• Performance-focused mindset
• Strong Python skills - experience with package management and Django
• Strong Golang ski