Software Engineer, Compiler (Front-end)

FuriosaAI
Seoul
Workplace: OnsiteFull timeFunction: Software EngineeringEducation: bachelorsSkills: ["Systems thinking","Extensibility design","Rigor","Clarity","Collaboration"]

Build the compiler front end that translates models from frameworks like PyTorch into FuriosaAI’s internal IR, then improves them with graph-level optimizations such as operator fusion, constant folding, and shape inference. Design an extensible ingestion system for new model architectures and custom operators, and develop/evolve a tensor-level kernel language and verification mechanisms to ensure correctness. Partner with software teams to improve end-to-end compilation and refine language design based on real usage.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
FuriosaAI
FuriosaAI
1 week ago

Software Engineer, Compiler (Front-end)

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 11 hours agoStatus: Live
Reposted: similar role first listed 3 months ago

Job Summary

Build the compiler front end that translates models from frameworks like PyTorch into FuriosaAI’s internal IR, then improves them with graph-level optimizations such as operator fusion, constant folding, and shape inference. Design an extensible ingestion system for new model architectures and custom operators, and develop/evolve a tensor-level kernel language and verification mechanisms to ensure correctness. Partner with software teams to improve end-to-end compilation and refine language design based on real usage.
Location: Seoul
Workplace: Onsite
Employment Type: Full time
Job Function: Software Engineering

Key Responsibilities

  • •Design and implement the front-end pipeline that transforms models from deep learning frameworks such as PyTorch into the compiler's internal IR.
  • •Develop graph-level optimizations including operator fusion, constant folding, shape inference, and layout transformations.
  • •Build extensible model ingestion structures to support new architectures (e.g., LLM, VLA, Multimodal) and custom operators while maintaining correctness.
  • •Design and evolve a tensor-level kernel language that exposes internal IR and DSL capabilities through a consistent user interface.
  • •Establish verification mechanisms for correctness and collaborate with software teams to refine compilation quality and language design.

Key Requirements

  • •Bachelor's degree in Computer Science, Mathematics, or a related field.
  • •Experience or familiarity with compilers, program transformation systems, or related infrastructure.
  • •Understanding of deep learning frameworks such as PyTorch, TensorFlow, and ONNX and their model representations.
  • •Ability to abstract complex system constraints into consistent, user-friendly programming interfaces.
  • •Proficiency in Python and experience with at least one systems programming language such as Rust or C++.
Experience:CompilersDeep learning frameworksAI accelerators
Education:Bachelor's in Computer Science, Mathematics, or related field
Skills:Systems thinkingExtensibility designRigorClarityCollaboration
Languages:English
Tech Stack:PythonRustC++PyTorchTensorFlowONNXTorchDynamoFX GraphTorch.compileTorch.exportTritonLLVMMLIRTVMNPUGPUTPU

Company Brief

FuriosaAI
Designs and develops data‑center AI inference accelerators (RNGD) and a full stack hardware‑software platform to deliver energy‑efficient, high‑performance AI compute for enterprise and cloud customers.
Industry: Hardware Devices
Company Size: Medium (51 to 250 employees)
Growth: Scaleup
Valuation: USD 500M to 1B
Funding: Series C
Headquarters: Seoul, South Korea
Founded: 2017
Glassdoor
Glassdoor: 4.6
WebsiteLinkedInGlassdoor