Machine Learning Engineer - New Grad

SambaNova Systems•San Jose, TX

About The Position

SambaNova is hiring a new graduate Machine Learning Engineer to train and evaluate models that make large language models less expensive to serve and more capable. The work may include training speculative decoding models and tuning their acceptance rates, compressing models while preserving accuracy, or designing and evaluating agentic workflows for tool use, planning, and long-horizon tasks. The Machine Learning team develops model-side techniques for SambaNova’s inference platform, including speculative decoding, quantization, sparsity, distillation, agentic systems, reinforcement learning, and post-training. The role spans research, applied science, and production, from designing novel approaches and prototypes to delivering improvements in products. The engineer will work with AI performance, compiler, runtime, and product teams to turn research results into measured end-to-end improvements.

Requirements

Bachelor’s degree in Computer Science, Electrical Engineering, Mathematics, or a related field.
Strong programming skills in Python.
Solid foundation in machine learning, linear algebra, probability, and numerical computing.
Hands-on experience with PyTorch, TensorFlow, or JAX.
Working knowledge of transformer architectures and the LLM training or inference pipeline.
Ability to design sound experiments, interpret results honestly, and communicate findings clearly.

Nice To Haves

Master’s or Ph.D. in Computer Science, Electrical Engineering, or a related field.
Research or internship experience in LLM training, post-training, or inference.
Experience with speculative decoding, distillation, quantization, sparsity, or other model-efficiency techniques.
Experience building agentic systems, tool-calling pipelines, or LLM evaluation harnesses.
Familiarity with reinforcement learning or preference optimization, including RLHF, DPO, GRPO, or reward modeling.
Experience with distributed training frameworks such as FSDP, DeepSpeed, or Megatron, or inference frameworks such as vLLM, SGLang, or TensorRT-LLM.
Publications at ML venues such as NeurIPS, ICML, ICLR, ACL, or EMNLP, or meaningful open-source contributions.

Responsibilities

Train, evaluate, and tune speculative decoding models, including Dspark/Dflash, Medusa/Eagle-style heads, and multi-token prediction, to increase acceptance rates on production workloads.
Develop and implement model optimization techniques such as quantization, sparsity, distillation, pruning, and parameter-efficient fine-tuning, and quantify trade-offs among model quality, latency, and resource efficiency.
Build and evaluate agentic systems for tool calling, multi-step planning, retrieval, and long-context reasoning on the serving stack.
Design rigorous evaluations and benchmarks, and substantiate accuracy and quality claims.
Translate recent research into working implementations and run ablation studies to determine what drives results.
Partner with AI performance, compiler, runtime, and product teams to ship model improvements into SambaNova products.
Contribute to reinforcement learning, post-training, and reward modeling as the roadmap expands.

Benefits

Competitive total rewards package including base salary, equity, and benefits.
95% premium coverage for employee medical insurance.
77% premium coverage for dependent medical insurance.
Health Savings Account (HSA) with employer contribution.
Dental, vision, short- and long-term disability, basic life, voluntary life, and AD&D insurance plans.
Flexible Spending Account (FSA) options for health care, limited purpose, and dependent care.
Headspace subscription.
Gympass+ membership with access to physical gyms.
One Medical membership.
Counseling services through an Employee Assistance Program.