2 months ago

Driver Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 19 hours agoStatus: Live

Job Summary

Build low-level driver infrastructure that sits between the compiler/runtime stack and accelerator silicon, enabling MAX and Mojo to run reliably and efficiently across NVIDIA, AMD, Apple Silicon, and new hardware. You’ll implement core driver abstractions (device, context, queue, memory, kernel launch), develop multi-accelerator/multi-node communication primitives, and improve async diagnostics for kernel and graph authors. Work across kernels, compiler/runtime, and serving on a fully open-source stack.

Location: United States, Canada

Workplace: Hybrid

Employment Type: Full time

Job Function: Transportation & Fleet Operations

Seniority: Mid level

Key Responsibilities

•Implement and extend core driver abstractions (Device, Context, Queue, Memory, Function/kernel representation and launch) across diverse hardware backends.
•Build multi-accelerator and multi-node communication and collectives primitives for large-model inference using technologies like NVLink, RDMA, and UCX.
•Improve diagnostics and error reporting across an asynchronous execution stack, converting low-level failures into actionable messages for production authors.
•Partner with Kernels, Graph Compiler/Runtime, and Serving teams to shape the surfaces used by Mojo standard library and MAX framework.
•Participate in design discussions and code reviews while contributing to a fully open-source project with public GitHub contributions.

Travel: Low travel

Pay and Benefits

Salary: USD 167,000 - 242,000 annually

Equity and Bonus:Equity

Perks:Health Insurance401kPaid Leave

Key Requirements

•3+ years writing high-performance, low-latency production systems in C++ (modern C++17/20), with strong ownership, lifetime, ABI stability, concurrency, and parallelism.
•Hands-on experience with at least one accelerator driver-level API such as CUDA Driver API, HIP, Metal, or Vulkan compute (streams, events, contexts, module loading).
•Working understanding of accelerator execution and memory models including stream ordering, host↔device transfers, and pinned memory.
•Solid instincts for library and API design, including naming, layering, and ergonomics.
•Strong debugging skills in asynchronous, multi-device systems using tools like GDB/LLDB and sanitizers.

Skills:OwnershipConcurrencyDebuggingCollaborationProblem-solving

Tech Stack:C++C++17C++20CUDA Driver APIHIPMetalVulkanStreamsEventsContextsModule loadingAsync allocatorsIPCNVLinkRDMAInfinibandRoCEEFASocketsUCX

Eligibility

Work Authorization:Authorization required. Sponsorship not provided.

Company Brief

Modular

Provides infrastructure and developer tools to build, deploy, and scale large AI models and foundation-model applications, including model hosting, orchestration, and SDKs to accelerate AI product development.

Industry: AI & Machine Learning

Website

Loading job details...

Preparing the role view and application actions.

Modular

2 months ago

Driver Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI

Last checked: 19 hours agoStatus: Live

Job Summary

Location: United States, Canada

Workplace: Hybrid

Employment Type: Full time

Job Function: Transportation & Fleet Operations

Seniority: Mid level

Key Responsibilities

•Implement and extend core driver abstractions (Device, Context, Queue, Memory, Function/kernel representation and launch) across diverse hardware backends.
•Build multi-accelerator and multi-node communication and collectives primitives for large-model inference using technologies like NVLink, RDMA, and UCX.
•Improve diagnostics and error reporting across an asynchronous execution stack, converting low-level failures into actionable messages for production authors.
•Partner with Kernels, Graph Compiler/Runtime, and Serving teams to shape the surfaces used by Mojo standard library and MAX framework.
•Participate in design discussions and code reviews while contributing to a fully open-source project with public GitHub contributions.

Travel: Low travel

Pay and Benefits

Salary: USD 167,000 - 242,000 annually

Equity and Bonus:Equity

Perks:Health Insurance401kPaid Leave

Key Requirements

•3+ years writing high-performance, low-latency production systems in C++ (modern C++17/20), with strong ownership, lifetime, ABI stability, concurrency, and parallelism.
•Hands-on experience with at least one accelerator driver-level API such as CUDA Driver API, HIP, Metal, or Vulkan compute (streams, events, contexts, module loading).
•Working understanding of accelerator execution and memory models including stream ordering, host↔device transfers, and pinned memory.
•Solid instincts for library and API design, including naming, layering, and ergonomics.
•Strong debugging skills in asynchronous, multi-device systems using tools like GDB/LLDB and sanitizers.

Skills:OwnershipConcurrencyDebuggingCollaborationProblem-solving

Tech Stack:C++C++17C++20CUDA Driver APIHIPMetalVulkanStreamsEventsContextsModule loadingAsync allocatorsIPCNVLinkRDMAInfinibandRoCEEFASocketsUCX

Eligibility

Work Authorization:Authorization required. Sponsorship not provided.

Company Brief

Modular

Industry: AI & Machine Learning

Website