AI/ML Compiler Lead Edge AI

Role & responsibilities

• Lead design and development of the AI/ML compiler toolchain for Ambients ultra-low power AI

chips.

• Build and optimize compiler backends, IR transformations, and runtime systems for AI workloads.
• Drive model conversion pipelines (TF, TFLite, PyTorch, ONNX Ambient compiler stack).
• Implement graph-level and operator-level optimizations (fusion, scheduling, memory reuse).
• Collaborate with hardware and data science teams to co-design hardware-aware compiler

strategies.

• Develop tools for quantization, pruning, mixed precision, and efficient memory allocation.
• Ensure compiler generates binaries with predictable latency, power, and memory footprint.
• Define performance metrics and benchmarking strategies for compiler outputs.
• Mentor a team of compiler and software engineers; review design, code, and optimization flows.
• Stay up to date with MLIR, LLVM, XLA and latest AI compiler trends.

Preferred candidate profile

• Bachelor’s/Master’s in CSE, ECE, EEE, or related.
• 7+ years of experience in compiler development, with at least 3+ years in ML compilers/AI

acceleration.

• Strong expertise in compiler internals: parsing, IR design, optimization passes, code generation.
• Proficiency in C/C++, Python, LLVM/MLIR, ONNX Runtime.
• Solid understanding of neural network architectures, graph optimization, and hardware

acceleration.

• Familiarity with DSPs, GPUs, and AI accelerators; knowledge of edge hardware (ARM Cortex-M,

RISC-V) desirable.

• Experience in runtime systems, scheduling, and memory management for constrained devices.
• Excellent leadership, problem-solving, and cross-functional collaboration skills.