Senior LLM Algorithm Developer (m/f/d)

Your mission

At Q.ANT, we are building photonic processing systems that compute with light-delivering a scalable,

energy-efficient alternative to transistor-based architectures for next-generation AI and HPC

(High-Performance Computing) applications. Large language models (LLMs) represent one of our prime workloads and our biggest opportunity.

As our Senior LLM Algorithm Developer, your mission is to rethink how LLMs compute from the ground up. Operating deep within the LLM architecture, you will explore how photonic compute capabilities can be best employed for AI workloads. This is a highly creative, research-driven role where your insights and inventions will directly shape our future hardware architecture, driving Q.ANT's ultimate success in the market.

Your Responsibilities

• Algorithmic Innovation: Rethink, design, and create new algorithms for LLMs tailored to photonic capabilities, optimizing existing algorithm layers to achieve breakthrough performance.
• Modeling & Simulation: Project, model, and simulate the performance and scalability of advanced LLM architectures to ensure future systems can leverage them with maximum confidence.
• Hardware-Software Co-Design: Inform and collaborate tightly with the PIC (Photonic Integrated Circuit) and Systems Architecture teams to define the design points of future hardware generations based on algorithmic requirements.
• Cross-Functional Collaboration: Act as the vital link bridging software and hardware architecture discussions, aligning closely with the overall vision and direction of the algorithms and applications team.
• Ownership & Initiative: Drive your research with a high degree of freedom, moving from first principles to concrete, inventive solutions that push the boundaries of what is possible in analog/digital electronics and photonics.

Your profile

• Education: University degree (MSc or PhD) in Computer Science, Mathematics, Physics, or a related technical field. A PhD focused on LLM core components is highly preferred.
• Technical Expertise: Deep algorithmic knowledge of how LLMs work and perform, including a comprehensive grasp of the bottlenecks and challenges of vanilla transformers.
• Hands-on Experience: Proven experience developing LLM core pieces. Deep academic/PhD research on these topics is highly valued (no formal industry experience required for PhD holders; MSc applicants should demonstrate relevant experience on top of their studies).
• Advanced Architectures (Bonus): Familiarity with hybrid architectures (e.g., State Space Models like Mamba and Gated DeltaNet), diffusion language models, reasoning models, or hardware-software co-design for AI accelerators is a strong advantage.
• Tools & Frameworks: High proficiency with PyTorch, JAX, frameworks like vllm, llama.cpp, or similar environments.
• Track Record: Significant contributions to LLM open-source projects or published papers/talks at leading AI/ML conferences are a huge plus.
• Mindset: Strong ownership, solution orientation, and "first principles" thinking. You thrive in a highly creative, research-centric environment where innovation is required.
• Communication: Fluent in English (working language).

Why us?

• We hire for attitude and train for skills. We value deep curiosity, a research-first mindset, and the drive to invent.
• We encourage self-responsibility and accelerate professional development. You will have the freedom to pioneer solutions at the absolute bleeding edge of tech.
• We move fast and encourage experimentation. Failure is just a data point on the road to a breakthrough.
• Unprecedented Impact: You will have the unique opportunity to develop breakthrough photonic computing technology and leave a lasting imprint on the future of AI infrastructure.
• World-Class Team: Work alongside a passionate, international, and highly skilled cross-functional team of physicists, hardware architects, and software engineers.