Multimodal Data Engine Expert

The role will focus on the architecture and development of an advanced multimodal data platform supporting heterogeneous computing, multimodal processing, vector storage, distributed caching, AI-enhanced database technologies and intelligent data agents.

You will help build end-to-end capabilities for processing and managing large-scale heterogeneous and unstructured data, covering compute scheduling, data processing, retrieval, storage and autonomous system management.

The objective is to develop high-performance data infrastructure optimised for large language models, multimodal AI workloads and emerging AI agent ecosystems., Multimodal Data Processing Platform

• Design and develop a unified platform for scheduling and managing heterogeneous compute resources including CPUs, GPUs and NPUs.
• Develop serverless resource pooling capabilities across multiple compute engines.
• Improve heterogeneous resource scheduling, workload placement and overall infrastructure utilisation.
• Support scalable deployment of data-processing workloads across distributed environments.

Multimodal Computing Engines

• Design and optimise processing engines for large-scale structured, unstructured and multimodal data.
• Support efficient processing of text, images, video and other data formats.
• Develop next-generation query optimisation and hybrid execution techniques.
• Enable high-performance retrieval and processing across vector, textual, geospatial and other data types.

Vector Storage and Indexing

• Design scalable systems for multimodal vector storage, retrieval and indexing.
• Develop capabilities for metadata management, access control and index lifecycle management.
• Integrate with relevant open-source data and AI ecosystems.
• Improve storage efficiency, data placement and retrieval performance for AI workloads.

Distributed Caching

• Design high-performance distributed caching services for multimodal data platforms.
• Provide low-latency near-compute caching for processing engines.
• Improve data movement and sharing between distributed compute engines.
• Optimise end-to-end performance across large-scale multimodal data pipelines.

AI-Powered Data Agents

• Apply AI4DB, large language models and agent technologies to develop intelligent data-management capabilities.
• Build agents and reusable skills that automate data processing, workload development, storage management, analytics and system operations.
• Explore autonomous optimisation and decision-making within database and data infrastructure.
• Integrate AI capabilities directly into data-management workflows.

• Strong programming skills in one or more languages such as C, C++, Python or Java.
• Strong technical background in areas such as:
• Database systems
• Big data platforms
• Distributed systems
• High-performance computing
• Practical or research experience working with core systems components such as:
• Query optimisers
• Query execution engines
• Storage engines
• Distributed storage systems
• Strong understanding of heterogeneous computing architectures involving CPUs, GPUs and NPUs.
• Experience with heterogeneous resource management, scheduling and performance optimisation.
• Ability to design systems that efficiently utilise large-scale compute resources.
• Experience developing, operating or maintaining cloud-computing platforms.
• Familiarity with modern DevOps and infrastructure engineering practices.

Preferred Experience

• Experience with LLM fine-tuning and reinforcement learning.
• Knowledge of multimodal AI technologies including:
• Natural language processing
• Computer vision
• Time-series processing
• Experience integrating AI technologies with database or data-management systems.
• Experience developing infrastructure for large-scale AI, multimodal or agent-based workloads.
• Strong understanding of performance engineering, scalability and distributed system design.