Inflection.xyz
Compiler Engineer, Parallel Computing Platforms
Inflection.xyz, Santa Clara, California, us, 95053
FABRIC CRYPTOGRAPHY
WE ARE BUILDING THE WORLD'S FIRST CRYPTOGRAPHIC COMPUTER
Fabric believes hardware determines the boundaries of humanity's collective creativity and imagination. We are building hardware for the next generation of cryptography because we believe in creating a more trustworthy world with secure, private computation at its core. Just as encryption and decryption enabled the Internet as we know it, this new paradigm of cryptographic algorithms, such as zero knowledge proofs, have even broader potential to revolutionize how trust, privacy, and identity work in our society.
About the job
We are seeking a Compiler Engineer with exceptional expertise in code generation for massively-parallel computing architectures. In this role, you will contribute to the design and implementation of advanced compiler technologies, enabling hardware-aware optimizations that unlock new levels of performance for our custom hardware.
What you'll be doing
Design and implement advanced compiler technologies, focusing on massively-parallel architectures and hardware-aware optimizations. Develop and extend MLIR (Multi-Level Intermediate Representation) compiler frameworks, including the creation of custom dialects and optimization passes. Optimize low-level code generation for instruction-level parallelism, memory access patterns, and efficient utilization of custom hardware resources. Collaborate with hardware architects to ensure seamless integration of compiler solutions with our hardware platforms. Lead efforts in hardware-aware compilation, leveraging deep knowledge of instruction set architectures and memory hierarchies. Innovate in domain-specific language (DSL) development and machine learning compiler frameworks as needed. Qualifications and experience requirements
Advanced understanding of MLIR's multi-level intermediate representation. Proven experience in designing and implementing custom dialects. Expert-level skills in compiler development, including code generation techniques, optimization passes, and low-level transformations. Fluency in systems programming languages such as C and C++. Deep understanding of massively-parallel computing platforms, including GPUs, distributed models, and parallelization techniques. Experience targeting hardware platforms like machine learning accelerators or custom AI hardware. Experience creating custom computational languages tailored to high-performance computing. Strong grasp of language design principles for specific domains. Familiarity with TensorFlow XLA, PyTorch compilation technologies, or similar frameworks. Experience optimizing computation graphs for machine learning. Fabric Cryptography is committed to fostering a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate based on race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law. #J-18808-Ljbffr
WE ARE BUILDING THE WORLD'S FIRST CRYPTOGRAPHIC COMPUTER
Fabric believes hardware determines the boundaries of humanity's collective creativity and imagination. We are building hardware for the next generation of cryptography because we believe in creating a more trustworthy world with secure, private computation at its core. Just as encryption and decryption enabled the Internet as we know it, this new paradigm of cryptographic algorithms, such as zero knowledge proofs, have even broader potential to revolutionize how trust, privacy, and identity work in our society.
About the job
We are seeking a Compiler Engineer with exceptional expertise in code generation for massively-parallel computing architectures. In this role, you will contribute to the design and implementation of advanced compiler technologies, enabling hardware-aware optimizations that unlock new levels of performance for our custom hardware.
What you'll be doing
Design and implement advanced compiler technologies, focusing on massively-parallel architectures and hardware-aware optimizations. Develop and extend MLIR (Multi-Level Intermediate Representation) compiler frameworks, including the creation of custom dialects and optimization passes. Optimize low-level code generation for instruction-level parallelism, memory access patterns, and efficient utilization of custom hardware resources. Collaborate with hardware architects to ensure seamless integration of compiler solutions with our hardware platforms. Lead efforts in hardware-aware compilation, leveraging deep knowledge of instruction set architectures and memory hierarchies. Innovate in domain-specific language (DSL) development and machine learning compiler frameworks as needed. Qualifications and experience requirements
Advanced understanding of MLIR's multi-level intermediate representation. Proven experience in designing and implementing custom dialects. Expert-level skills in compiler development, including code generation techniques, optimization passes, and low-level transformations. Fluency in systems programming languages such as C and C++. Deep understanding of massively-parallel computing platforms, including GPUs, distributed models, and parallelization techniques. Experience targeting hardware platforms like machine learning accelerators or custom AI hardware. Experience creating custom computational languages tailored to high-performance computing. Strong grasp of language design principles for specific domains. Familiarity with TensorFlow XLA, PyTorch compilation technologies, or similar frameworks. Experience optimizing computation graphs for machine learning. Fabric Cryptography is committed to fostering a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate based on race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law. #J-18808-Ljbffr