Tesla

Makes electric cars, robotaxis and grid batteries

Machine Learning Engineer - Model Quantization

Applied Machine LearningConfirmed live in the last 24 hours

Full-timePalo Alto, CA, USANo salary listedIn-person

About the Job

Requirements

A degree or equivalent experience in Computer Science, Machine Learning, Robotics, Computer Vision, or a related quantitative field.
At least 2 years of hands-on experience training, optimizing, and deploying large-scale quantized deep learning models.
Strong technical understanding of quantizing large transformer architectures, including mitigating activation outliers, KV cache quantization, and maintaining numerical stability of low-precision attention mechanisms.
Deep expertise in the theory and low-level implementation of modern quantization algorithms such as GPTQ, AWQ, SmoothQuant, and OmniQuant.
Experience with low-level numerics and emerging data formats such as FP8, INT4, W4A8, W8A8, and micro-scaling/MX formats, and their trade-offs regarding latency, memory bandwidth, and model fidelity.
Rigorous understanding of computer architecture and the roofline model, including familiarity with optimizing memory hierarchies, minimizing SRAM/DRAM data movement, and mapping quantized GEMMs and memory-bound operators to custom silicon.
Proficiency in writing custom CUDA/Triton kernels, implementing custom autograd functions such as Straight-Through Estimators, and manipulating PyTorch computational graphs such as FX tracing and torch.compile.
Strong software engineering skills, including writing clean, production-grade Python and C++ code that ships reliably at scale.
Proven ability to turn cutting-edge research into robust, real-world systems that improve safety, capability, efficiency, or digital productivity.
Passion for Tesla’s mission and excitement about deploying SI that moves both the physical and digital worlds forward.

Responsibilities

Architect and scale Post-Training Quantization and Quantization-Aware Training pipelines for massive multimodal foundation models combining vision, prediction, and decision-making, optimizing inference latency, memory bandwidth utilization, and power consumption for self-driving cars, Optimus robots, and digital agents.
Innovate quantization-aware-training recipes and algorithms to address optimization challenges inherent to low-precision training.
Research and implement advanced low-bit-weight post-training quantization techniques addressing activation outlier mitigation, KV cache compression, and layer-wise bit allocation while maintaining model accuracy.
Collaborate with SI compiler, inference engine, and silicon teams to co-design quantization-friendly architectures, hardware-aware sparsity patterns, and mixed low-precision kernels that use hardware capabilities effectively.
Collaborate across perception, planning, robotics, digital agents, and infrastructure teams to move models from research into fleet-wide, robot-wide, and enterprise-wide deployment.

Category	Tesla SI
Location	Palo Alto, California
Requisition ID	271231
Type	Full-time

Job Description

At Tesla SI, we’re not just training models, we’re building the foundation models that power the future of real-world autonomy. Your work will directly control millions of Tesla vehicles and robotaxis on the road today, Optimus humanoid robots in factories and homes, and Digital Optimus — our groundbreaking SI agents that autonomously operate computers and software systems at enterprise scale. In addition, we deploy these models to edge hardware where efficiency and accuracy is paramount.Whether your background is in pushing the limits of ultra-low-bit precision, developing post-training quantization and quantization-aware-training algorithms, or designing model architecture for quantized inference, if you excel at making massive deep learning models run lightning-fast on custom silicon and want to see your research deployed at planetary scale, this role is built for you.At Tesla SI we celebrate and enable speed, ownership, and real-world impact. Join our SI Team and you’ll have: Unparalleled real-world data (tens of billions of miles of driving + robot interactions + digital workflows); One of the largest SI training clusters on Earth; Immediate closed-loop feedback from vehicles, physical robots, and real-time computer interfaces; The ability to ship improvements to millions of customers, robots, and digital agents in weeks, not years.If you want your models to solve real physics, causality, long-horizon planning, dexterous manipulation, and autonomous digital task execution, this is the highest-leverage SI role on the planet. Let’s build the future together.

Job Responsibilities

Architect and scale quantization pipelines (both Post-Training Quantization and Quantization-Aware Training) for massive multi-modal foundation models that fuse vision, prediction, and decision-making. You will optimize inference latency, memory bandwidth utilization, and power consumption for self-driving cars, Optimus robots, and digital agents operating at enterprise scale
Innovate quantization-aware-training recipes and algorithms that tackle complex optimization challenges inherent to low-precision training
Push the limits of low-precision AI: Research and implement advanced low-bitweight post-training quantization techniques to address hard algorithmic problems such as activation outlier mitigation, KV cache compression, and optimal layer-wise bit-allocation while strictly maintaining model accuracy
Collaborate closely with SI compiler, inference engine, and silicon teams to ensure models are architected to maximally utilize underlying hardware capabilities by co-designing quantization-friendly architectures, hardware-aware sparsity patterns, and mixed low-precision kernels
Collaborate across perception, planning, robotics, digital agents, and infrastructure teams to move models from research to fleet-wide, robot-wide, and enterprise-wide deployment

Job Requirements

Degree or equivalent experience in Computer Science, Machine Learning, Robotics, Computer Vision, or related quantitative field
2+ years of hands-on experience training, optimizing, and deploying large-scale quantized deep learning models
Strong technical understanding of the challenges inherent to quantizing large transformer architectures, including mitigating massive activation outliers, KV cache quantization, and maintaining the numerical stability of attention mechanisms at low precision
Deep expertise in the theory and low-level implementation of modern quantization algorithms (e.g., GPTQ, AWQ, SmoothQuant, OmniQuant)
Experience with low-level numerics and emerging data formats (e.g., FP8, INT4, W4A8, W8A8, micro-scaling/MX formats) and their trade-offs regarding latency, memory bandwidth, and model fidelity
Rigorous understanding of computer architecture and the roofline model. Familiarity with how to optimize for memory hierarchies, minimize SRAM/DRAM data movement, and efficiently map quantized GEMMs and memory-bound operators to custom silicon
Proficiency in writing custom CUDA/Triton kernels, implementing custom autograd functions (e.g., Straight-Through Estimators), and manipulating PyTorch computational graphs (e.g., FX tracing, torch.compile)
Strong software engineering skills — clean, production-grade Python/C++ code that ships reliably at scale
Proven ability to turn cutting-edge research into robust, real-world systems that improve safety, capability, efficiency, or digital productivity
Passion for Tesla’s mission and excitement about deploying SI that moves both the physical and digital worlds forward

About the Company

Tesla

Makes electric cars, robotaxis and grid batteries

Tesla makes electric cars and batteries that store power for homes and utilities, and it develops self-driving software and humanoid robots. In 2023, its Model Y became the first electric vehicle to rank as the world's best-selling car. Buyers order the Model 3 sedan, Model Y SUV, or Cybertruck pickup directly from Tesla instead of through dealers, and they charge on its Supercharger fast-charging network. It also sells Megapack and Powerwall batteries, runs a robotaxi service, and charges drivers a monthly fee for self-driving software that still needs their supervision. Tesla was founded in 2003, with Elon Musk as its lead early investor and a recognized co-founder, and it is based in Austin, Texas.

Company size: 10,001+
Company stage: IPO
Headquarters: Austin, TX
Founded: 2003

Recently Posted Jobs

Explore more opportunities at Tesla

See more jobs