Software Engineer, Compiler (Front-end)

Job Title

Software Engineer, Compiler (Front-end)

Role Summary

Design and implement the front-end of a production compiler for AI accelerators, responsible for model ingestion, graph-level transformations, and a tensor-level kernel language. Work within the compiler team to ensure correct, extensible translations from external frameworks into the compiler's internal IR and improve end-to-end compilation quality.

Experience Level

Mid-level. The posting does not specify years of experience but expects prior practical experience with compilers and deep learning frameworks.

Responsibilities

Primary responsibilities include building the front-end pipeline, implementing graph optimizations, and designing a user-facing kernel language.

• Implement model ingestion from frameworks such as PyTorch into the compiler internal IR.
• Develop graph-level optimizations: operator fusion, constant folding, shape inference, and layout transforms.
• Build extensible ingestion structures to support LLMs, VLA, multimodal models, and custom operators.
• Design and evolve a tensor-level kernel language that maps to the internal IR and maximizes hardware performance.
• Establish verification and validation mechanisms to ensure translation correctness.
• Collaborate with software teams and users to refine language design and improve compilation quality based on real usage.

Requirements

Must-have technical skills and experience for day-one contribution.

• Experience or familiarity with compilers, program transformation systems, or related infrastructure.
• Understanding of deep learning frameworks and their model representations (PyTorch, TensorFlow, ONNX).
• Ability to abstract system constraints into consistent, user-friendly programming interfaces.
• Proficiency in Python and experience with at least one systems language such as Rust or C++.

Nice-to-have

• Experience designing and implementing domain-specific languages or user-facing programming models.
• Knowledge of PyTorch compiler internals (TorchDynamo, FX Graph, torch.compile, torch.export) or kernel languages like Triton.
• Experience with compiler frameworks such as LLVM, MLIR, or TVM.
• Understanding of AI accelerator architectures (NPU, GPU, TPU) and graph-level compilation optimizations.
• Research or open-source contributions in compilers or deep learning frameworks.

Education Requirements

Minimum: Bachelor's degree in Computer Science, Mathematics, or a related field. Preferred: Master’s or PhD in Programming Languages, Compilers, Program Analysis, or related fields.

About the Company

Company: FuriosaAI

Headquarters: Seoul, South Korea

FuriosaAI develops high-performance, energy-efficient AI inference hardware and software. Founded in 2017 by semiconductor and AI engineers, the company builds AI-native compute platforms to reduce AI energy and operational costs and operates globally with offices in Korea, Silicon Valley, and an R&D lab in Lisbon.