AI Performance Engineer - New Grad

$131k - $197k

Location
Austin, Texas, United States
Workplace
On-site
Employment
Full-time

About this role

SambaNova is a leader in next-generation AI infrastructure, delivering a full-stack inference platform for customers worldwide. At the core of SambaNova's technology is the RDU (Reconfigurable Dataflow Unit) — a chip built on a dataflow architecture rather than the traditional GPU model. Its decode performance is especially strong for agentic workloads like multi-turn agents, code generation, and long-running applications. RDUs are packaged into SambaRack, rack-scale hardware that lets customers deploy state-of-the-art models with better performance, greater energy efficiency, and faster time to value.

About the Team

The AI Performance team brings up and optimizes foundation models such as DeepSeek, Llama, Qwen, and GPT OSS on SambaNova's dataflow platform. We profile and tune models across the compiler, runtime, and hardware layers to improve inference throughput and latency.

About the Role

As a new grad AI Performance Engineer, you'll bring up new open source models on the SambaNova stack and work to make them run faster. You'll profile model execution, find bottlenecks, and apply optimizations such as quantization, kernel fusion, and improved scheduling with the compiler, runtime, and hardware teams.

Responsibilities

  • Bring up new foundation models on the SambaNova platform using our software stack
  • Profile model performance across the compiler, runtime, and hardware layers
  • Identify bottlenecks and propose dataflow or scheduling optimizations for single-node and distributed systems
  • Apply techniques from recent research in quantization, scheduling, and memory optimization
  • Benchmark throughput and latency and track performance changes across releases
  • Work with the machine learning, compiler, runtime, and hardware teams to ship performance improvements into SambaNova products

Required Qualifications

  • B.S. in Computer Science, Electrical Engineering, or related field
  • Strong programming skills in Python or C++
  • Understanding of algorithms, data structures, and numerical computing
  • Experience with PyTorch, TensorFlow, or JAX
  • Experience with inference frameworks such as vLLM, TensorRT, or SGLang
  • GPU programming experience with CUDA or Triton

Preferred Qualifications

  • M.S. or Ph.D. in Computer Science, Electrical Engineering, or related field
  • Experience with LLM training or inference
  • Familiarity with quantization, graph optimization, kernel fusion, or model partitioning
  • Knowledge of memory hierarchy, caching, and scheduling for large model execution
  • Internship or research experience in ML systems or performance optimization
  • Publications or open source contributions in ML systems

Base Salary Range:

Base Pay Range
$131,200—$196,800 USD

Submission Guidelines
Please note that in order to be considered an applicant for any position at SambaNova Systems, you must submit an application form for each position for which you believe you are qualified. 

EEO Policy
SambaNova Systems is an Equal Opportunity/Affirmative Action Employer. All qualified applicants will receive consideration for employment without regard basis of age (40 and over), color, disability, gender identity, genetic information, marital status, military or veteran status, national origin/ancestry, race, religion, creed, sex (including pregnancy, childbirth, breastfeeding), sexual orientation, and any other applicable status protected by federal, state, or local laws.

Benefits Summary for US-Based, Full-Time Employment Positions
SambaNova offers a competitive total rewards package, including the base salary, plus equity and benefits. We cover 95% premium coverage for employee medical insurance, and 77% premium coverage for dependents and offer a Health Savings Account (HSA) with employer contribution. We also offer Dental, Vision, Short/Long term Disability, Basic Life, Voluntary Life, and AD&D insurance plans in addition to Flexible Spending Account (FSA) options like Health Care, Limited Purpose, and Dependent Care. Our library of well-being benefits available to you and your dependents includes a full subscription to Headspace, Gympass+ membership with access to physical gyms, One Medical membership, counseling services with an Employee Assistance Program, and much more.

Apply to SambaNova

Clera fills in SambaNova's application for you.

How it works

Open the role in Clera

Sign in, and this role opens on your Clera job board, ready to apply.

Clera fills in the application

Clera fills the form from your profile and resume, and asks what's missing.

Track it in Clera

Each application you send shows up in your Clera tracker, sorted by stage.

Want Clera to find roles for you?

Ask your AI about this role

It knows you. Clera gives it the role.

Does this role fit my profile?

First time? Connect Clera once, about a minute.
One click above pre-fills the connector. Or by hand: Settings > Connectors > Add custom connector, paste https://mcp.getclera.com. Either way, sign in with the account you use on Clera and approve the consent screen.
https://mcp.getclera.com

Sign in with your Clera account. MCP docs

About SambaNova

Welcome to SambaNova: Revolutionizing AI Capacity

SambaNova, co-founded by Rodrigo Liang (CEO), Kunle Olukotun (Chief Technologist), and Chris Re, empowers developers, enterprises, governments, and data centers to unlock their full AI potential. Founded in 2017 in Silicon Valley, our full-stack infrastructure, from chips to models, enables lightning-fast performance, low power consumption, and high-efficiency computing.

Our Mission

We give every developer, enterprise, government and data center absolute sovereignty over their own data, models and AI infrastructure. SambaNova future-proofs the AI workloads that will power and scale the businesses of tomorrow.

The Speed That Matters is Payback
SambaNova delivers premium inference. Anyone can be fast. The question is whether you make money doing it. Customers make more money by buying SambaNova chips.

SambaNova AI chips are unmatched — paying for themselves in 6 months or less. When a company purchases a rack, they’ll see payback in half a year. Not only does the rack draw less power because of how the chip is built, but it’s also air-cooled — and can be installed in a data center the customer already owns. This reduces time and data center buildout, while being energy-efficient and sustainable. The racks are flexible, plugging in from cell towers to a retrofitted CPU hall. SambaNova racks earn money in three ways: how fast tokens are made, what each token sells for, and how many customers it can serve at once.

Most organizations want to know, “How long will it take to get my investment back?” We’ve cut down the ROI timeline. Paybacks from other companies can take up to 24 months. We’re at six months on a machine that lasts 5 years — leaving most of its life as profit.

Our Tech

We provide our customers the opportunity to experience SambaNova technology through the cloud or on-premise through full-stack AI platforms, AI racks, and AI chips with SambaCloud, SambaStack, SambaRack, and Reconfigurable Dataflow Unit.

Team size
201-500 employees people
Founded
2017
Industry
Computer Hardware Manufacturing
Workplace
On-site
Locations
San Jose, California, United States

Benefits

  • Medical insurance
  • Dental insurance
  • Vision insurance
  • Health savings account
  • Short term disability
  • Long term disability
  • Basic life insurance
  • Voluntary life insurance
  • AD&D insurance
  • Flexible spending account
  • Headspace subscription
  • Gympass membership
  • One medical membership
  • Employee assistance program
  • Equity