G

Genesis AI

Develops general-purpose robots with AI stack

Compiler Lead

Systems Software EngineeringConfirmed live in the last 24 hours
Full-timeParis, FranceNo salary listedHybrid

About the Job

Requirements

  • Expertise in GPU optimization, including access patterns, register and shared-memory usage, barriers, atomics, and threading; CPU intuition is a plus.
  • Idiomatic Python skills and a strong understanding of API design for Python users.
  • Knowledge of compiler pipelines, including intermediate representations, lowering, and optimization.
  • Ability to reason about whole-system architecture and where complexity should live.
  • Ability to measure performance before optimizing, including memory traffic and launch overhead.
  • Ability to write tests and benchmarks alongside implementation.
  • Proactive, self-directed working style, with the ability to gather context across the company.
  • Comfort communicating with engineers who use the software and designing in the open for external users.

Responsibilities

  • Own Quadrants end to end, including its API, intermediate representation and lowering, optimization passes, numerical primitives, and roadmap.
  • Design the authoring experience for physics and numerical-computing engineers, including decorators, constraints, error handling, and escape hatches.
  • Maintain and extend the intermediate representation, lower it to multiple backends, and preserve room for differentiable simulation.
  • Optimize execution plans through vectorization, register and shared-memory usage, barriers and atomics, cooperative threading, dead-buffer and redundant-traffic elimination, and removal of unnecessary synchronization.
  • Develop and own numerical primitives such as Cholesky, eigendecomposition, matrix multiplication, reductions, and solver kernels, including correctness, benchmarks, and regression tracking.
  • Find performance improvements in both Genesis and Quadrants, and propose API changes that reduce duplication, improve readability, and reduce platform-specific special-casing.
  • Benchmark real workloads, track performance over time, catch regressions early, engage with users, identify friction and missing abstractions, drive adoption, and set direction.
  • Keep Quadrants healthy and unblock the company while building the open-source Python compiler for physics simulation.
  • Read and challenge the codebase, reduce technical debt, maintain a sound core, monitor competing technologies, and incorporate and ship relevant advances.

Desired Qualifications

  • CPU intuition.

About the Company

Genesis AI

Develops general-purpose robots with AI stack

Genesis AI designs and builds general-purpose robots for industrial use by developing an end-to-end in-house robotics stack. Its system combines a robot brain called GENE for complex manipulation, a dexterous robot hand with a data-collection glove to learn from human motion, and Genesis World, a simulation platform for training in virtual scenarios. The real robot, Eno, emphasizes practical capability and reliability for mass production, with feedback from deployments continually updating models, simulations, and hardware. The company targets manufacturing, logistics, and laboratories first, with plans to expand into hospitality, healthcare, consumer products, and outdoor applications.

Company size
51-200
Company stage
Series A
Total funding
$605M
Headquarters
Paris, France
Founded
2024

Recently Posted Jobs

Explore more opportunities at Genesis AI

See more jobs