HPC Engineer

Graviton Research Capital runs quantitative trading strategies built on time series analysis, filtering, classification, stochastic models and pattern recognition over terabytes of market data. That research needs serious compute, and this role runs it. You deploy and manage high performance computing clusters, work with cluster management tooling and large scale batch scheduling systems such as PBS Pro, Torque or Slurm, and handle testing, user support and root cause analysis when jobs or nodes misbehave. Parallel filesystem environments are useful, as is comfort with multiprocessor and multithreading concepts. The stack is Linux with Python and Bash. Graviton names 2 to 3 years as the ideal, and says explicitly that Bachelor's graduates will be considered with an exceptional background and prior HPC experience, so the door is open slightly below that range.

What the posting asks for: the ideal candidate has 2 to 3 years of experience, with a degree in EE, CS or CSE. Graviton adds that Bachelor's graduates will also be considered given an exceptional background and prior experience in the HPC field. Required skills are programming in Python or Bash, a good working understanding of Linux at user level, experience with large scale distributed HPC systems, and an understanding of large scale batch processing systems such as PBS Pro, Torque or Slurm along with approaches to efficient scheduling. HPC system monitoring and automation tools are preferred.

The responsibilities as listed: hands on experience deploying high performance clusters, working with cluster management tools, exposure to multiprocessor and multithreading concepts, working on Linux, and handling testing, user support, troubleshooting and root cause analysis of issues. Experience with parallel filesystem environments is marked optional.

Location and setup: Gurugram, on site. Published benefits are competitive compensation, an annual international team outing, fully covered commuting expenses, health insurance, catered breakfast and lunch, a stocked kitchen, 4 weeks of annual leave plus market holidays, gym and sports club memberships, and regular social events.

Honest fit guidance: this is infrastructure and systems work, not trading or model development, and the day to day is closer to a research computing role at a university or national lab than to a software product team. That makes it an unusual and quite specific opening. The explicit note that strong Bachelor's graduates will be considered is worth taking seriously if you have real cluster experience from an academic or internship setting, since HPC exposure is genuinely hard to come by in India outside a handful of employers. Note this is Graviton's third open role today alongside a Python engineer role and an application reliability role. The posting does not publish an interview process.