At Hybridizer, we are redefining performance portability. Our flagship technology, Hybridizer, allows developers to compile high-level bytecode (like .NET and Java) into highly optimized source code for multicore CPUs and GPUs. We help our clients achieve hardware-peak performance without the complexity of rewriting their entire codebases. We are looking for a brilliant Senior Engineer to join our team in Paris to tackle the most challenging problems in high-performance computing.
Your MissionYour primary goal is to bridge the gap between high-level logic and bare-metal performance. You will not just write code; you will think, implement, and build the engines that drive complex algorithms for our most demanding customers.
Think: Analyze specific customer challenges in finance, simulation, or data processing. Deconstruct their mathematical models and determine how to map them efficiently onto parallel architectures using our compiler technology.
Implement: Architect and develop robust solutions that extend the capabilities of the Hybridizer. You will work deep within the compilation stack, ensuring that abstract instructions translate into optimized machine code (CUDA/ROCm) that screams performance.
Build: Create complex, reusable algorithms and libraries. You will take ownership of critical components of our technology, optimizing for throughput and latency, and proving that our generated code matches or beats hand-tuned binaries.
You are a problem solver who loves the intersection of mathematics, hardware, and software engineering. You understand what happens "under the hood" when a line of code executes.
Expert GPU Programming: You have deep, hands-on experience with CUDA or ROCm. You understand memory hierarchies, warp divergence, and how to squeeze every FLOP out of a GPU.
Compiler Knowledge: You are familiar with how compilers work (ASTs, IR, optimization passes). Experience with LLVM or writing transpilers is a massive plus.
Mathematical Rigor: You possess strong math problem-solving skills. You are comfortable diving into complex algorithms (linear algebra, stencils, Monte Carlo simulations) and restructuring them for parallel execution.
High-Level Engineering: While you know the low-level details, you are also a seasoned software engineer capable of writing clean, maintainable, and architecturally sound code in C++, C#, or Java.
Customer Focus: You can translate a client's "slow calculation" problem into a concrete engineering roadmap and deliver a solution that works.
At Hybridizer, we offer a collaborative and innovative work environment where your contributions make a real impact. Join us to work on challenging projects, grow your skills, and be part of a team that values creativity and excellence.
Skills Required
- Deep hands-on experience with CUDA or ROCm GPU programming
- Understanding of GPU memory hierarchies, warp divergence, and performance optimization
- Familiarity with compiler architecture, including ASTs, intermediate representations, and optimization passes
- Strong mathematical problem-solving skills involving algorithms such as linear algebra, stencils, or Monte Carlo simulations
- Seasoned software engineering experience with C++, C#, or Java
- Experience with LLVM or writing transpilers
- Ability to translate customer performance problems into engineering solutions
What We Do
Hybridizer is a software platform for performance portability and GPU acceleration. Its compiler transforms C#/.NET and Java bytecode or high-level code into optimized source code for multicore CPUs and GPUs, allowing developers to use existing codebases without learning CUDA or rewriting applications. The technology supports debugging, profiling, cross-platform deployment, and demanding workloads such as quantitative finance, scientific simulation, and data processing.






