AI Compiler Engineer

Overview

AI Compiler Engineer to develop and optimize graph compilers for AI/ML workloads and enable efficient model deployment on EnCharge’s inference accelerators.

What you'll do

• Architect, design, and implement AI model execution optimizations in graph compilers to improve performance and reduce latency.
• Collaborate with ML researchers, hardware engineers, and software developers to design and deploy AI models on hardware.
• Optimize neural network execution using techniques such as layer fusion, operator fusion, and graph-level transformations.
• Build compiler passes that convert high-level models (e.g., TensorFlow, PyTorch) into intermediate representations (IR).
• Implement parsing, semantic analysis, and IR generation for deep learning frameworks.
• Research and integrate compiler design and ML/hardware acceleration advancements into graph compilers.
• Provide leadership, mentorship, and technical guidance to engineers working on graph compiler optimizations.

What you'll need

• 3+ years of compiler development focused on AI/ML graph compilers.
• Proficiency with AI graph compiler frameworks such as MLIR and Torch-FX.
• Solid background in hardware architectures such as GPUs, TPUs, and ASICs.
• Experience with optimization techniques including fusion, quantization, and tiling.
• Strong knowledge of intermediate representations, code parsing, and semantic analysis.
• Proficiency in C++ and/or Python (or other compiler-development languages).

Details

• Location: India.