ML Compiler Architect, Senior Principal at d-Matrix

Toronto, Ontario, Canada

Apply Now

Not SpecifiedCompensation

Senior (5 to 8 years), Expert & Leadership (9+ years)Experience Level

Full TimeJob Type

UnknownVisa

Artificial Intelligence, SemiconductorsIndustries

Requirements

BS with 15+ years / MS with 12+ years / PhD with 10+ years in Computer Science or Electrical Engineering
12+ years of experience in Front End Compiler and systems software development, focused on ML inference
Deep experience in designing or leading compiler efforts using MLIR, LLVM, Torch-MLIR, or similar frameworks
Strong understanding of model optimization for inference: quantization, fusion, tensor layout transformation, memory hierarchy utilization, and scheduling
Expertise in deploying ML models to heterogeneous compute environments, with attention to latency, throughput, and resource scaling in cloud systems
Proven track record working with AI frameworks (e.g., PyTorch, TensorFlow), ONNX, and hardware backends
Experience with cloud infrastructure, including resource provisioning, distributed execution, and profiling tools
Preferred Qualifications
Experience targeting inference accelerators (AI ASICs, FPGAs, GPUs) in cloud-scale environments

Responsibilities

Architect the MLIR-based compiler for cloud inference workloads, focusing on efficient mapping of large-scale AI models (e.g., LLMs, Transformers, Torch-MLIR) onto distributed compute and memory hierarchies
Lead the development of compiler passes for model partitioning, operator fusion, tensor layout optimization, memory tiling, and latency-aware scheduling
Design support for hybrid offline/online compilation and deployment flows with runtime-aware mapping, allowing for adaptive resource utilization and load balancing in cloud scenarios
Define compiler abstractions that interoperate efficiently with runtime systems, orchestration layers, and cloud deployment frameworks
Drive scalability, reproducibility, and performance through well-designed IR transformations and distributed execution strategies
Mentor and guide a team of compiler engineers to deliver high-performance inference-optimized software stacks

Skills

MLIR

LLVM

Compiler Architecture

AI Inference

Transformer Models

NLP

Heterogeneous Compute

Dynamic Partitioning

Workload Scheduling

Cloud Deployment

Tensor Processors

Vector Engines

d-Matrix

AI compute platform for datacenters

About d-Matrix

d-Matrix focuses on improving the efficiency of AI computing for large datacenter customers. Its main product is the digital in-memory compute (DIMC) engine, which combines computing capabilities directly within programmable memory. This design helps reduce power consumption and enhances data processing speed while ensuring accuracy. d-Matrix differentiates itself from competitors by offering a modular and scalable approach, utilizing low-power chiplets that can be tailored for different applications. The company's goal is to provide high-performance, energy-efficient AI inference solutions to large-scale datacenter operators.

Santa Clara, CaliforniaHeadquarters

2019Year Founded

$149.8MTotal Funding

SERIES_BCompany Stage

Enterprise Software, AI & Machine LearningIndustries

201-500Employees