Talent.com
CUDA Kernel Engineer
CUDA Kernel EngineerPRAGMATIKE • San Francisco, California, United States
CUDA Kernel Engineer

CUDA Kernel Engineer

PRAGMATIKE • San Francisco, California, United States
[job_card.variable_days_ago]
[job_preview.job_type]
  • [job_card.full_time]
  • [filters_job_card.quick_apply]
[job_card.job_description]

Location : Cambridge, MA (Eastern Time / UTC -4) Relocation package available

Start date : ASAP

Languages : English (required)

About the Role

Pragmatike is hiring on behalf of a fast-growing AI startup recognized as a Top 10 GenAI company by GTM Capital , founded by MIT CSAIL researchers.

We are searching for a CUDA Kernel Engineer who has hands-on experience developing and optimizing NVIDIA CUDA kernels from scratch . You will work on the GPU performance layer powering large-scale, high-throughput AI systems used by Fortune 500 customers.

This role is ideal for someone who deeply understands NVIDIA GPU architecture, memory hierarchy, warp-level execution, and profiling workflows not someone coming from generic hardware, FPGA, or non-NVIDIA compute backgrounds. You will directly influence the GPU efficiency, throughput, and scalability of mission-critical AI systems.

What Youll Do

  • Design, implement, and optimize custom CUDA kernels for NVIDIA GPUs , with a focus on maximizing occupancy, memory throughput, and warp efficiency.
  • Profile GPU workloads using tools such as N sight Compute, Nsight Systems, nvprof, and CUDA‐MEMCHECK .
  • Analyze and eliminate performance bottlenecks including warp divergence, uncoalesced memory access, register pressure, and PCIe transfer overhead.
  • Improve GPU memory pipelines (global, shared, L2, texture memory) and ensure proper memory coalescing.
  • Collaborate closely with AI systems, model acceleration, and backend distributed systems teams.
  • Contribute to GPU architecture decisions, kernel libraries, and internal performance-engineering best practices.

What Were Looking For

  • Proven track record building NVIDIA CUDA kernels from scratchnot just calling existing libraries.
  • Strong ability to optimize kernels (tiling strategies, occupancy tuning, shared memory design, warp scheduling).
  • Deep understanding of CUDA threads, warps, blocks, and grids, GPU memory hierarchy and memory coalescing, as well as warp divergence (how to detect, analyze, and mitigate it)
  • Experience diagnosing PCIe bottlenecks and optimizing host-device transfers (pinned memory, streams, batching, overlap).
  • Familiarity with C++, CUDA runtime APIs, and GPU debugging / profiling tooling.
  • Bonus Points

  • Experience with multi-GPU or distributed GPU systems (NCCL, NVLink, MIG).
  • Background in GPU acceleration for ML frameworks or HPC workloads.
  • Knowledge of model inference optimization (TensorRT, CUDA Graphs, CUTLASS).
  • Exposure to compiler-level optimization or PTX / SASS analysis.
  • Startup experience or comfort working in fast-moving, ambiguous environments.
  • Why This Role Will Pivot Your Career

  • Research pedigree : MIT CSAIL founders recognized for breakthrough AI and systems contributions.
  • Customer impact : Deploy AI solutions powering Fortune 500 clients .
  • Industry momentum : Lab alumni have led high-value acquisitions (MosaicML Databricks, Run : AI Nvidia, W&B CoreWeave).
  • Funding & growth : Oversubscribed seed round, next funding in 2026.
  • Career growth & influence : Lead AI initiatives, optimize pipelines, and directly impact production AI systems at scale .
  • Culture & autonomy : Own critical systems while collaborating with world-class engineers.
  • Aspirational impact : Solve GPU / AI performance challenges few engineers ever face.
  • Benefits

  • Competitive salary & equity options
  • Sign-on bonus
  • Health, Dental, and Vision
  • 401k
  • Pragmatike is an Equal Opportunity Employer and is committed to providing equal employment opportunities to all applicants without discrimination. We recruit on behalf of our clients and prohibit discrimination and harassment based on race, color, religion, age, sex, national origin, disability status, genetics, protected veteran status, sexual orientation, gender identity or expression, or any other characteristic protected by federal, state, or local laws. This policy applies to all terms and conditions of employment, including recruiting, hiring, placement, promotion, termination, layoff, recall, transfer, leaves of absence, compensation, and training.We are committed to a fair and inclusive hiring process. We process your personal data solely for recruitment purposes, in accordance with applicable privacy laws, and maintain reasonable safeguards to protect your information. Your data may be shared with our client(s) for hiring consideration, but will not be disclosed to third parties outside of the recruitment process.

    [job_alerts.create_a_job]

    Engineer • San Francisco, California, United States

    [internal_linking.similar_jobs]
    GPU Systems Engineer - HPC / Parallel Computing

    GPU Systems Engineer - HPC / Parallel Computing

    Vast.ai • San Francisco, CA, US
    [job_card.full_time]
    AI projects and businesses all over the world.We are democratizing and decentralizing AI computing—reshaping our future for the benefit of humanity. We are a small, growing, and highly motivat...[show_more]
    [last_updated.last_updated_30] • [promoted]
    Software Engineer, Mobile

    Software Engineer, Mobile

    Gridware • San Francisco, CA, US
    [job_card.full_time]
    Gridware is a San Francisco-based technology company dedicated to protecting and enhancing the electrical grid.We pioneered a groundbreaking new class of grid management called active grid response...[show_more]
    [last_updated.last_updated_30] • [promoted]
    Software Engineer - C++ Core Middleware

    Software Engineer - C++ Core Middleware

    Zoox • Foster City, CA, US
    [job_card.full_time]
    Shipping the software that powers Zoox's revolutionary autonomous vehicles is an extremely challenging job.As a Software Engineer on the Robotics Middleware team, you will be pivotal in designi...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    Senior Backend Engineer (Go / C++)

    Senior Backend Engineer (Go / C++)

    Suki • Redwood City, California, United States
    [job_card.full_time]
    The Future of Healthcare Needs You.At Suki, we’re building technology that listens, understands, and gets out of the way — so clinicians can get back to being clinicians. AI to automate clinical doc...[show_more]
    [last_updated.last_updated_30] • [promoted]
    Performance ML Engineer : CUDA, GPU Systems

    Performance ML Engineer : CUDA, GPU Systems

    Relace • San Francisco, CA, United States
    [job_card.full_time]
    A tech company specializing in ML infrastructure is seeking a Machine Learning Engineer who excels at making models faster and more efficient through performance tuning and optimization.The ideal c...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    Engineer II - IAM Technologies

    Engineer II - IAM Technologies

    Exelixis • Alameda, CA, United States
    [job_card.full_time]
    The Engineer II - Client Technology provides advanced engineering support across a broad range of technologies and platforms. This role plays a critical part in anticipating and resolving escalated ...[show_more]
    [last_updated.last_updated_30] • [promoted]
    GPU Systems Engineer

    GPU Systems Engineer

    SLR Search • San Francisco, CA, US
    [job_card.full_time]
    Architect the foundation of the future's most performance-critical cloud infrastructure.Starting Salary targeting $200,000 - $280,000. Comprehensive medical, dental, vision.GPU Systems Engineer ...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    AI Kernel Engineer

    AI Kernel Engineer

    Quadric, Inc • Burlingame, California, United States
    [job_card.full_time]
    Quadric has created an innovative general purpose neural processing unit (GPNPU) architecture.Quadric's co-optimized software and hardware is targeted to run neural network (NN) inference workloads...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    Full-Stack Engineer

    Full-Stack Engineer

    Renegade • San Francisco, CA, US
    [job_card.full_time]
    Renegade is building an unstoppable network for the anonymous exchange of value.Our core permissionless protocol, the Renegade dark pool, solves many problems in current decentralized exchange desi...[show_more]
    [last_updated.last_updated_30] • [promoted]
    ML Engineer

    ML Engineer

    Wispr Flow • San Francisco, California, United States
    [job_card.full_time]
    Wispr Flow is making it as effortless to interact with your devices as talking to a close friend.Voice is the most natural, powerful way to communicate — and we’re building the interfaces to make t...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    Backend Engineer

    Backend Engineer

    Layer • San Francisco, California, United States
    [job_card.full_time]
    We want engineers who can build a team and a company, not just a best-in-class product.We believe that certainly takes broad technical competence across our tech stack, but it also takes harder to ...[show_more]
    [last_updated.last_updated_30] • [promoted]
    Chroma - Backend Engineer

    Chroma - Backend Engineer

    Ai Talent Flow • San Francisco, California, United States
    [job_card.full_time]
    Scaling retrieval systems is a challenging engineering problem.Chroma's distributed data plane must support workloads appropriate to serving AI applications at scale, which involves scaling vector ...[show_more]
    [last_updated.last_updated_30] • [promoted]
    Senior Engineer, Pega

    Senior Engineer, Pega

    VirtualVocations • Oakland, California, United States
    [job_card.full_time]
    A company is looking for a Senior Engineer, Pega / Constellation.Key Responsibilities Contribute to project success by delivering medium-to-large features and reducing technical debt Develop comp...[show_more]
    [last_updated.last_updated_variable_hours] • [promoted] • [new]
    Sustaining Engineer II

    Sustaining Engineer II

    Jupiter Endovascular • Menlo Park, CA, US
    [job_card.full_time]
    Imagine a world where you could bring the precision and control of surgery to catheter-based therapies.For decades, endovascular procedures have been constrained by the technological limitations of...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    Production Engineer, Compute

    Production Engineer, Compute

    Crusoe • San Francisco, CA, US
    [job_card.full_time]
    Crusoe's mission is to accelerate the abundance of energy and intelligence.We’re crafting the engine that powers a world where people can create ambitiously with AI — without sacrif...[show_more]
    [last_updated.last_updated_30] • [promoted]
    GPU Fleet Engineer – Hyperscale Infra, Kubernetes & AI

    GPU Fleet Engineer – Hyperscale Infra, Kubernetes & AI

    OpenAI • San Francisco, CA, United States
    [job_card.full_time]
    Join a forward-thinking company as an engineer in the fleet infrastructure team, where you'll design and operate systems for one of the largest GPU fleets globally. This role offers the chance to wo...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]
    Backend Engineer, Models

    Backend Engineer, Models

    Meter • San Francisco, California, United States
    [job_card.full_time]
    Networking is one of the most fundamental industries in all of technology.For the first time, Meter has unified the full networking stack. and now we are making it autonomous.Meter Zero is our neur...[show_more]
    [last_updated.last_updated_30] • [promoted]
    Backend Engineer

    Backend Engineer

    Strala Group • San Francisco, California, United States
    [job_card.full_time]
    Strala is completely rebuilding the foundation of the world’s largest industry - Insurance Claims Processing.In less than six months, we reached mid six-figure ARR and raised a 13x oversubscribed $...[show_more]
    [last_updated.last_updated_variable_days] • [promoted]