Research Engineer, AI Models

Location: Germany

Salary: €116,000–€154,000 EUR per year

Company

Building an AI platform based on in-memory computing architecture designed to improve energy efficiency and performance for AI inference workloads.

What you will do

• Research and implement techniques for accelerating AI inference, including quantization, sparsity, distillation, speculative decoding, caching strategies, and architectural modifications.
• Characterize tradeoffs between model quality, latency, throughput, and power consumption across use cases.
• Partner with hardware, compiler, and quantization teams to translate algorithmic improvements into gains on custom silicon.
• Build profiling tools and benchmarking frameworks to identify bottlenecks and measure model quality and efficiency.
• Develop fine-tuning and post-training workflows using LoRA, adapters, and full fine-tuning for modern AI models.
• Evaluate new architectures and techniques and contribute research insights to technical and go-to-market strategy.

Requirements

• 5+ years of experience in ML research, applied ML, or ML systems.
• Strong Python and PyTorch fundamentals.
• Hands-on experience with transformers, diffusion models, or state space models.
• Experience fine-tuning large models and building training and evaluation pipelines.
• Deep understanding of transformers, attention mechanisms, and optimization techniques.
• Ability to read and implement techniques from research papers.

Nice to have

• Experience with efficient inference techniques such as KV cache optimization, attention variants, MoE routing, or flow matching.
• Background in hardware-aware ML optimization or quantization.
• Familiarity with PyTorch Profiler, Nsight, or custom instrumentation.
• Publications in generative modeling, efficient inference, or ML systems.
• Contributions to open-source ML projects.

Culture & Benefits

• Work at the intersection of ML research, software engineering, and AI hardware.
• Contribute to a vertically integrated AI stack spanning models, compilers, systems, and silicon.
• Collaborate with experienced AI researchers, silicon and systems engineers, and architects.