Senior Researcher – LLM System Architecture

Senior Researcher – LLM Systems & Inference Architecture

Position Overview

A research-driven technology organization is seeking a Senior Researcher specializing in large language model (LLM) systems and inference architecture.

This role focuses on advancing the efficiency, scalability, and performance of LLM inference systems through system-level optimization and hardware–software co-design. The position operates at the intersection of AI systems, compilers, and heterogeneous computing platforms.

Key Responsibilities

LLM Inference Engine Optimization

• Design and implement high-performance inference engines for large language models.

• Improve efficiency of both open-source and internal inference frameworks.

• Develop optimization techniques such as:

  • Model quantization

  • Sparse attention mechanisms

  • Key-value cache reuse and memory optimization

• Enhance throughput and latency for transformer-based and generative AI workloads.

Hardware–Software Co-Design

• Design and implement optimized compute kernels for heterogeneous accelerators.

• Enable efficient execution of LLM workloads across specialized hardware platforms.

• Profile end-to-end AI workflows to identify system bottlenecks across runtime, framework, and hardware layers.

• Develop low-latency, high-throughput solutions bridging Python-based frameworks and accelerator backends.

Performance Engineering & System Optimization

• Analyze and optimize large-scale AI pipelines, focusing on:

  • Memory bandwidth utilization

  • Compute efficiency

  • Scheduling and execution pipelines

• Improve system-level performance across distributed or heterogeneous environments.

Research & Ecosystem Contribution

• Publish research in leading systems and machine learning conferences.

• Contribute to open-source AI systems and tooling ecosystems.

• Support adoption of AI infrastructure through:

  • Developer tools

  • Documentation

  • Collaboration with external partners

Required Qualifications

• PhD or Master’s degree in Computer Science, Computer Engineering, or a related field.

• Strong programming skills in Python and C/C++ for system-level development.

• Solid experience with AI frameworks such as PyTorch or TensorFlow.

• Deep understanding of model optimization techniques for large-scale neural networks.

• Experience with system-level performance profiling and optimization.

Preferred Qualifications

• Experience with heterogeneous or accelerator programming environments (e.g., GPU, NPU, or other xPU architectures).

• Familiarity with kernel development, custom operators, or compiler-level optimization.

• Experience with large multimodal or vision-language models.

• Contributions to open-source AI infrastructure or research publications in top-tier venues.

Personal Attributes

• Strong systems thinking with ability to bridge AI models and hardware platforms.

• Research-oriented mindset combined with practical engineering skills.

• Ability to work independently while collaborating across multidisciplinary teams.

• Clear technical communication skills.