Argus Array Data Production Engineer

The Argus Array will be the largest optical telescope array ever assembled, with a collecting area comparable to the largest monolithic telescopes in the world.  The Data Production Engineer will bring the real-time analysis pipeline into production, working with a team of scientists to develop, implement, and deploy cutting edge algorithms and approaches to astronomical difference image analysis, image resampling and reprojection, and coaddition. You will write and design code that interfaces directly with the largest digital sensor array ever assembled — a distributed machine vision camera with more pixels than 6000 4K TVs. You will work with a cross-functional team responsible for the entire data path of the Argus Array, from the telescope telemetry and hardware operation to long-term data archiving and distribution.

Minimum Education and Experience Requirements

Relevant post-Baccalaureate degree required (or foreign degree equivalent); for candidates demonstrating comparable independent research productivity or professional-level background in sponsored research administration, will accept a relevant Bachelor’s degree (or foreign degree equivalent) and 3 or more years of relevant experience in substitution.

Required Qualifications, Competencies, and Experience

• We are searching for an engineer with deep Python expertise (5+ years) and demonstrated success in designing and optimizing high-throughput systems (2+ years). 
• Experience and proficiency across the software development lifecycle (version control, documentation, and testing) is required.
• Experience with GPU acceleration frameworks (Nvidia CUDA, PyCUDA, CuPy, or equivalent) is required.

Preferred Qualifications

• Experience with maturing research-grade code to production in machine vision, astronomical data processing, or similar field is preferred. 
• Prior experience translating scientific requirements into technical specifications and working in a mixed research environment is preferred. 
• Experience with cluster-scale and distributed computing in a DevOps context is strongly preferred, including working with standard frameworks (Ray, Spark, Dask, or equivalent).

Ability to sustain nighttime validations testing and monitoring of your deployed data pipelines (as part of a rota of qualified support personnel).