LLVM / MLIR Compiler Architect - Photonic Computing

Q.ant Gmbh
München, Germany
26 days ago

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Experience level
Expert
Experience required
5 years minimum
Working hours
Regular working hours
Languages
English, German
Job source

Tech stack

Artificial Intelligence Systems Engineering Computer Engineering Field-Programmable Gate Array (FPGA) Hardware Design Hardware Interface Design OpenMP Open Source Technology Tensorflow Scientific Computating Systems Architecture Graphics Processing Unit (GPU)
+9 more
Application Specific Integrated Circuits Pytorch Parallel Computation Build Management Information Technology Instruction Set Architecture Low Latency Free and Open-Source Software Programming Languages

Job description

The future of AI computing builds upon light. Q.ANT is building photonic processing systems that compute with light - delivering a scalable, energy-efficient alternative to transistor-based architectures for next-generation AI and HPC applications.

As LLVM / MLIR Compiler Architect - Photonic Computing (f/m/d), you will design and build the compiler stack that brings photonic computing to life. You will shape the interface between developers and hardware, define how software frameworks drive photonic execution, and co-design the system architecture that will power Q.ANT’s next-generation products.

This isn’t incremental optimization - it’s defining the compiler structure from the ground up for a fundamentally new compute paradigm. You’ll sit at the boundary to algorithms and software ecosystems, and you feed into low level software interfaces and hardware architecture. Your design decisions will shape how Q.ANT’s photonic processors evolve across generations and how efficiently they execute the next generation of AI workloads.

  • Design, lead, and deliver the Q.ANT compiler stack - shape the LLVM/MLIR-based compilation framework for photonic computing
  • Define the MLIR dialect for Q.ANT and build the path into the hardware interfaces and co-design the system architecture and Instruction Set Architecture (ISA)
  • Shape the hardware design point to ensure large parts of algorithms can execute on-device with high parallelism, low latency, and maximum utilization
  • Provide technical guidance to algorithm and application teams - help internal teams develop and optimize high-performance compute kernels for photonic execution
  • Drive profiling and performance analysis - analyze compiler-generated code to identify bottlenecks and optimization opportunities
  • Lead collaboration with external partners and internal teams (system engineering, architecture, ASIC design) - coordinate compiler stack development while ensuring cohesive design across the hardware/software stack
  • Contribute to Q.ANT’s Open Source initiatives - engage with the LLVM/MLIR community and Open Source development practices

Requirements

  • 5+ years building compilers (LLVM/MLIR) and co-designing hardware architectures, including profiling and performance analysis of compiler-generated code
  • Deep expertise in systems architecture, programming languages, and AI/scientific computing algorithms - you understand how compute architectures work and what algorithms need from hardware
  • Master’s degree or PhD in Computer Science, Computer Engineering, or related field; equivalent industry experience considered
  • Fluent in English; German language beneficial
  • Experience with novel or heterogeneous compute architectures (GPUs, TPUs, FPGAs, custom accelerators) is a plus
  • Hands-on experience defining MLIR dialects and contributing to LLVM/MLIR Open Source projects is a plus
  • Experience with AI frameworks (PyTorch, JAX, TensorFlow) and parallel programming (OpenMP, MPI) is a plus
  • Background in photonics is beneficial (but not required)

About the company

  • Make impact at scale: Help solve one of computing’s biggest challenges - making growing demand in compute and sustainability go hand in hand
  • Work on the leading edge: Photonic AI acceleration technologies that will define the next decade of computing - already validated in production HPC environments
  • Own your work from day one: Shape design and code of software, client solutions, and public collaterals on your own, with direct impact on product success and the future of AI infrastructure
  • Fast-track your growth: Work on challenges with few industry precedents - every problem you solve becomes new institutional knowledge and potentially industry-defining IP
  • Be the multiplier: Your work enables entire developer communities and directly helps clients succeed with photonic computing
  • World-class team: Collaborate alongside a passionate, international, cross-functional team of experts in photonics, processor design, and AI software
  • Direct access to leadership: Work closely with the company’s founders, including CEO Dr. Michael Förtsch, and the Q.ANT senior leadership team
  • Collaborative culture: Innovative work environment that values technical excellence, open communication, and pragmatic problem-solving
  • Be part of history: Join at the inflection point where photonic computing transitions from research to mainstream - your contributions will shape this transformation, Q.ANT is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees.

Apply for this position

This job is hosted externally. Click below to view the full posting and apply.

Apply on www.xing.com

Good distractions

Talks and stories from around this role — technically off-topic, practically not.

1:39 min

Fundamentals of tensors and the TensorFlow library

Håkan Silfvernagel · LIVE

4:37 min

Simplifying parallel programming with the CUDA ecosystem

Paul Graham Paul Graham · LIVE

2:35 min

Preventing remote code execution in PyTorch models

Balázs Kiss · WWC 2023

1:48 min

Compiling quantum software architectures using machine learning integrations

Matic Petrič Matic Petrič · WWC 2025

3:55 min

Evaluating central server APIs against edge deployment models

Hauke Brammer · WWC 2021

1:06 min

Compiling PyTorch environments for advanced time forecasting

Christoph Lohrmann Christoph Lohrmann +1 · WWC Europe 2026

Videos

See all

Related articles

See all