Requirements

• PhD in computer science, machine learning, ML systems, computer systems, computer architecture, electrical engineering, applied mathematics, or a closely related discipline.
• Strong publication record or equivalent research artifacts in ML, ML systems, efficient inference, model compression, quantization, distillation, serving systems, or related areas.
• Strong Python and PyTorch implementation skills with the ability to create experiments and working prototypes.
• Deep knowledge of LLMs, VLMs, transformer inference, decoding algorithms, model compression, quantization, and production-serving trade-offs.
• Strong experimental design skills including ablations, baselines, metrics, statistical reasoning, and failure analysis.
• Excellent written and verbal communication skills.

Nice to haves

• First-author publications at major venues like NeurIPS, ICML, ICLR, MLSys, ACL, EMNLP, ASPLOS, OSDI, SOSP, ISCA, or HPCA.
• Experience deploying ML models or inference optimizations in production environments.
• Familiarity with tools/technologies such as vLLM, SGLang, TensorRT-LLM, NVIDIA Dynamo, FlashAttention, FlashInfer, Triton, CUDA, or PyTorch internals.
• Experience in applying post-training, SFT, DPO, RLHF, RLAIF, preference optimization, or synthetic data generation for inference quality or efficiency.
• Contributions to open-source research artifacts, widely used benchmarks, technical blogs, or invited talks related to efficient AI systems.
• Research or implementation experience in distributed inference architectures including LLM request routing, PDD, multi-node serving, MoE expert parallelism, or WideEP.

What you'll be doing

• Lead research projects focused on efficient LLM and VLM inference, from hypothesis development to production handoff.
• Develop and evaluate techniques such as quantization, QAT, distillation, speculative decoding, KV-cache reuse/compression, long-context inference, MoE routing, and model/runtime co-optimization.
• Build prototypes using PyTorch, Triton, CUDA-adjacent tooling, or inference-serving frameworks and collaborate with engineering teams to deliver production components.
• Research request routing and scheduling strategies including cache-aware load balancing and prefill–decode disaggregation (PDD) to optimize latency, throughput, and cost.
• Investigate multi-node inference for dense and mixture-of-experts models with focus on expert placement, load balancing, and communication trade-offs.
• Conduct rigorous evaluations covering model quality, latency, throughput, numerical stability, memory footprint, tail latency, and cost per token.
• Collaborate with cross-functional teams to prioritize research with measurable production impact.
• Share findings through reports, blogs, papers, open-source artifacts, and mentor colleagues on experimental design and scientific rigor.

Perks and benefits

• Competitive compensation packages.
• Career growth and learning opportunities.
• Flexibility and ownership in your work.
• Collaborative and innovative work culture.
• Opportunity to contribute to impactful AI projects.
• International environment with talented teams.