Conference Description
Key Takeaways
- International academic conference dedicated to Big Data Engineering research and industrial applications
- Special focus on scalable inference and data engineering for large language models in production environments
- Designed for researchers, PhD students, postdoctoral fellows, and industry practitioners
- Features peer-reviewed paper presentations, keynote speeches, and specialised technical sessions
- Hosted at Keio University in Yokohama, Japan, from 5–7 August 2026
Introduction
The 8th International Conference on Big Data Engineering (BDE 2026) convenes researchers, engineers, and industry practitioners working at the intersection of large-scale data systems and modern artificial intelligence infrastructure. Scheduled for August 2026 at Keio University in Yokohama, Japan, the conference addresses one of the most pressing technical challenges facing organisations today: how to engineer data systems capable of supporting increasingly complex AI workloads at scale.
The timing is particularly significant as enterprises across sectors grapple with the operational demands of deploying large language models in production. These systems require fundamentally different approaches to data pipelines, inference infrastructure, and resource management compared to traditional analytics workloads.
About This Event
BDE 2026 represents the eighth iteration of this international forum, which has established itself as a venue for sharing peer-reviewed research alongside practical implementation experiences. The three-day programme combines keynote presentations from recognised experts with paper sessions covering both theoretical advances and real-world applications.
The conference maintains a strong emphasis on early-career researchers, actively encouraging participation from PhD students and postdoctoral fellows. This approach fosters knowledge transfer between established practitioners and emerging talent while creating opportunities for cross-institutional collaboration. Proceedings from the conference are published through ACM, providing authors with formal recognition within the academic community.
Scalable Inference and Production Data Engineering
A central theme of BDE 2026 concerns the engineering challenges associated with deploying large language models at scale. While much attention in the AI field focuses on model architecture and training methodologies, production deployment introduces distinct technical problems that fall squarely within the domain of data engineering.
Scalable inference—the ability to serve model predictions efficiently across millions of requests—requires careful consideration of data preprocessing pipelines, caching strategies, and computational resource allocation. These challenges intensify as organisations move beyond experimental deployments toward systems that must operate reliably under variable load conditions.
The conference dedicates special sessions to these topics, recognising that the gap between research prototypes and production-ready systems often represents the most significant barrier to realising value from AI investments.
Bridging Academic Research and Industrial Practice
Big Data Engineering occupies a distinctive position at the boundary between computer science research and operational technology practice. Academic contributions advance theoretical understanding of distributed systems, data structures, and algorithmic efficiency, while industrial practitioners contribute hard-won insights about what actually works when systems must operate continuously at scale.
BDE 2026 explicitly cultivates this exchange by bringing together participants from universities, research institutions, and technology companies. The resulting discussions often surface practical constraints that shape research directions while simultaneously exposing practitioners to emerging techniques not yet widely adopted in industry.
Who Should Attend
The conference serves professionals engaged with large-scale data systems across both academic and commercial contexts. Data engineers and data scientists working on production infrastructure will find relevant content addressing operational challenges, while research scientists and academic faculty can engage with the latest peer-reviewed work in the field.
Software development engineers building data-intensive applications, particularly those incorporating machine learning components, represent another core constituency. The emphasis on scalable inference makes the programme especially relevant for teams responsible for AI infrastructure within their organisations.
Venue and Academic Support
Keio University serves as both host institution and academic supporter through its Keio_AI initiative. Located in Yokohama, the venue provides access to one of Japan’s leading research universities while situating attendees within the broader Tokyo metropolitan area’s technology ecosystem. The in-person format facilitates the informal exchanges and relationship-building that remain difficult to replicate in virtual settings.