Modular

1 month ago

Driver Engineer

✓ Verified Job

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

Source: Company careers pageValidated by: Fursa AI

Last checked: 8 hours agoStatus: Live

Job Summary

Drive the core driver Abstractions between Modular's compiler/runtime and silicon, enabling reliable cross-backend execution on NVLink, RDMA, and various accelerators. Collaborate with Kernels, Graph Compiler/Runtime, and Serving teams to refine surfaces in Mojo and MAX, contribute to open source, and ensure high performance, low-latency execution across US/Canada environments.

Location: United States

Workplace: Hybrid

Employment Type: Full time

Job Function: Transportation & Fleet Operations

Key Responsibilities

•Implement and extend core driver abstractions — Device, Context, Queue, Memory, Function (kernel representation and launch) — across diverse hardware backends, including bindings that expose vendor-specific functionality where it matters.
•Contribute to multi-accelerator and multi-node communication and collectives primitives underpinning large-model inference, using NVLink, RDMA, and sockets (via UCX).
•Improve diagnostics and error reporting across the asynchronous execution stack, turning low-level driver failures into actionable messages for kernel/graph authors.
•Partner with Kernels, Graph Compiler/Runtime, and Serving teams to build and refine surfaces in the Mojo standard library and MAX framework.
•Participate in design discussions, code reviews, and collaborative software development to uphold high engineering standards.

Pay and Benefits

Salary: USD 167,000 - 242,000 annually

Equity and Bonus:Equity

Perks:401kEquityPaid Leave

Key Requirements

•3+ years of experience writing high-performance, low-latency production systems in C++ (modern C++17/20) with ownership, lifetime, ABI stability, concurrency, and parallelism.
•Hands-on experience with at least one accelerator driver-level API (CUDA Driver API, HIP, Metal, or Vulkan) including streams, events, contexts, and module loading.
•Working understanding of accelerator execution and memory models (stream ordering, host↔device transfers, pinned memory) with exposure to multi-GPU topologies (NVLink/xGMI, NUMA).
•Strong library and API design instincts — naming, layering, ergonomics.
•Strong debugging skills in asynchronous, multi-device systems (GDB/LLDB, sanitizers, CUDA/HIP error modes, race conditions, leaked resources)

Experience:3+ yearsAI infrastructureSystems programmingGPU acceleration

Skills:CollaborationCommunicationProblem-solvingAttention to detailProactive

Tech Stack:C++CUDACUDA Driver APIHIPMetalVulkanUCXNVLinkRDMAInfinibandRoCEEFAGDBLLDBSanitizers

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

1 month ago

Driver Engineer

✓ Verified Job

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

Source: Company careers pageValidated by: Fursa AI

Last checked: 8 hours agoStatus: Live

Job Summary

Location: United States

Workplace: Hybrid

Employment Type: Full time

Job Function: Transportation & Fleet Operations

Key Responsibilities

•Implement and extend core driver abstractions — Device, Context, Queue, Memory, Function (kernel representation and launch) — across diverse hardware backends, including bindings that expose vendor-specific functionality where it matters.
•Contribute to multi-accelerator and multi-node communication and collectives primitives underpinning large-model inference, using NVLink, RDMA, and sockets (via UCX).
•Improve diagnostics and error reporting across the asynchronous execution stack, turning low-level driver failures into actionable messages for kernel/graph authors.
•Partner with Kernels, Graph Compiler/Runtime, and Serving teams to build and refine surfaces in the Mojo standard library and MAX framework.
•Participate in design discussions, code reviews, and collaborative software development to uphold high engineering standards.

Pay and Benefits

Salary: USD 167,000 - 242,000 annually

Equity and Bonus:Equity

Perks:401kEquityPaid Leave

Key Requirements

•3+ years of experience writing high-performance, low-latency production systems in C++ (modern C++17/20) with ownership, lifetime, ABI stability, concurrency, and parallelism.
•Hands-on experience with at least one accelerator driver-level API (CUDA Driver API, HIP, Metal, or Vulkan) including streams, events, contexts, and module loading.
•Working understanding of accelerator execution and memory models (stream ordering, host↔device transfers, pinned memory) with exposure to multi-GPU topologies (NVLink/xGMI, NUMA).
•Strong library and API design instincts — naming, layering, ergonomics.
•Strong debugging skills in asynchronous, multi-device systems (GDB/LLDB, sanitizers, CUDA/HIP error modes, race conditions, leaked resources)

Experience:3+ yearsAI infrastructureSystems programmingGPU acceleration

Skills:CollaborationCommunicationProblem-solvingAttention to detailProactive

Tech Stack:C++CUDACUDA Driver APIHIPMetalVulkanUCXNVLinkRDMAInfinibandRoCEEFAGDBLLDBSanitizers

Company Brief

Modular

Industry: AI & Machine Learning

Website