Princeton University Hiring for Performance Tuning Analyst at Nassau Hall Full Time

Do you want to help enable and advance groundbreaking academic research? Do you have a background in research, writing code, and love to optimize code for advanced computing and storage systems? Do you want to support the use and development of research software in a centralized high-performance computing environment?  

Princeton Research Computing, a consortium of campus groups led by the Princeton Institute for Computational Science & Engineering (PICSciE) and the Office of Information Technology’s Research Computing group, is recruiting a full-time research software performance tuning analyst to join its team.  

In this position, you will be an integral member of the Researcher Engagement group as you collaborate with researchers from across multiple domains to accelerate active research code and workflows. This involves meeting with research groups to understand goals and current limitations, identify code or research workflow practices that are in need of tuning, profile the code or analyze the workflow, and identify bottlenecks. With the results of this process, you’ll work both independently and in partnership with researchers to develop new research workflows, new algorithms, improve parallelization and/or vectorization, and ultimately reduce the time to solution. In many cases, the performance improvements you make will enable researchers working on high-profile projects to produce new results or tackle larger problems. In doing this work, you will be responsible for monitoring, routing, and managing support requests submitted using our general support ticketing system, but with a particular focus on code or workflow improvement opportunities. You will work in close partnership with Research Computing and other IT colleagues, faculty, staff, and students to enable and accelerate their research computing efforts. This position reports to the Director of Researcher Engagement. 

The Researcher Engagement group is committed to creating a collaborative environment in which researchers from all disciplines at any experience or skill level are supported by computational and data science expertise, where best practices are valued, and where sharing and applying cross-disciplinary techniques to new and emerging areas is prioritized. 

Responsibilities

• Assist students, staff, and faculty with parallelizing, debugging, porting, benchmarking, and tuning research computing code 
• Provide intermediate and advanced support to Princeton University affiliated researchers in general programming (e.g., C/C++, MATLAB, Fortran, Python, R, Julia), parallel programming (e.g., MPI, OpenMP, multiprocessing in Python), debugging (e.g., GDB, DDT), use of CUDA and the NVIDIA HPC SDK, and a variety of machine learning tools (e.g., TensorFlow, PyTorch, Keras, Scikit-learn) 
• Support and troubleshoot users’ efficient use of our HPC clusters, installation of software, configuring or optimizing user-installed software, and management of data 
• Using a tool named Request Tracker, monitor and manage help desk tickets to solve research computing problems with special emphasis on workflow and code optimization 
• Participate in our twice weekly, one-hour research computing and programming support sessions that are open to the campus research community 
• Actively expand your technical skill set and gain experience in the use of software development tools and techniques, best practices, programming languages, high-performance computing hardware, and computational research solutions 
• Lead workshops, demonstrations, or other learning opportunities available to students, staff, and faculty. 
• Provide technical support during scheduled workshops and training sessions. 

Qualifications

Essential Qualifications

• 3 or more years of experience with scientific software development, including tuning and optimizing code, in a Linux environment 
• Experience working in an academic research environment 
• Strong programming skills, particularly with compiled and high-level scripting languages used in an HPC environment 
• Parallel programming experience using MPI and OpenMP 
• Experience using a job scheduling system (SLURM, LSF, PBS, etc.)
• Experience with programming for GPU accelerators using CUDA, OpenACC, OpenMP and/or OpenCL 
• Experience with tuning and debugging software used in HPC application development. For example, DDT, TotalView, VTune, Advisor, NVProf, MAP, TAU, Perf, etc. 
• Strong knowledge of at least one development tool and the desire to learn from others 
• Programming experience in a Red Hat Linux (or derivative such as CentOS) environment 
• Strong knowledge of development tools, including debuggers and compilers  
• Strong interpersonal skills and ability to respond promptly and professionally to requests for assistance from faculty, staff and students 
• Strong verbal and written communication skills 
• Ability to work and efficiently manage projects independently 
• Ability to learn new programming languages and technologies beyond the area of core knowledge 
• Ability to communicate effectively with a diverse user base having varied levels of technical proficiencies. 
• Education: A bachelor’s degree in computer science, engineering, sciences, or a related field is required. 

Princeton University is an Equal Opportunity/Affirmative Action Employer and all qualified applicants will receive consideration for employment without regard to age, race, color, religion, sex, sexual orientation, gender identity or expression, national origin, disability status, protected veteran status, or any other characteristic protected by law. KNOW YOUR RIGHTS

Standard Weekly Hours36.25Eligible for OvertimeNoBenefits EligibleYesProbationary Period180 daysEssential Services Personnel (see policy for detail)NoPhysical Capacity Exam RequiredNoValid Driver’s License RequiredNo Experience LevelMid-Senior Level#Ll-DP