Research Scientist-Model Efficiency (Intern)

About the Company

Bitdeer is a world-leading technology company for Bitcoin mining and AI cloud. Bitdeer provides comprehensive Bitcoin mining solutions, including designing industry-leading ASIC chips, manufacturing mining rigs, and managing complex processes across the value chain such as equipment procurement, transport logistics, datacenter design and construction, equipment management, and network and facility operations. The company also offers advanced cloud capabilities for artificial intelligence applications.

Headquartered in Singapore, Bitdeer operates globally with a diversified 3 GW energy portfolio and deploys Bitcoin mining and HPC datacenters in the United States, Bhutan, Norway, Canada, Malaysia, and Ethiopia.

About Bitdeer AI Lab

Bitdeer AI Lab is a frontier AI lab under Bitdeer, focused on exploring the frontiers of artificial intelligence with a long-term vision. The lab’s mission is to turn energy into affordable intelligence, emphasizing the economics of serving models to determine what technologies get built. The team works from the ground up, managing power, datacenters, and software to optimize AI infrastructure.

Responsibilities

This role focuses on making models cheaper and faster to serve without sacrificing quality. You will:

• Implement and adapt published methods such as quantization, sparsity and pruning, speculative decoding, MTP, and serving-time attention and KV-cache optimizations on Bitdeer’s models and hardware.
• Develop custom optimizations where existing methods fall short.
• Build rigorous evaluation frameworks to defensibly measure improvements (e.g., "lossless at 2× throughput").
• Own and present a well-defined 3–6 month project that delivers a tangible result.

Qualifications

• Undergraduate, Master's, or PhD candidates in Computer Science, Electrical Engineering, Mathematics, or related fields.
• Able to commit 3–6 months (full-time preferred).
• Strong programming ability in Python and hands-on experience with PyTorch.
• Coursework, self-study, or research experience in model efficiency techniques (e.g., quantization, sparsity, speculative decoding, or serving-time attention).
• Depth in at least one project, paper, or open-source contribution that you can explain end-to-end, including challenges faced.
• Solid understanding of transformer internals and a habit of rigorous evaluation (e.g., task-level metrics, controlled comparisons, honest baselines).
• Proven achievements in academics, engineering projects, open-source contributions, or programming/algorithm competitions are highly preferred.
• Deep enthusiasm for cutting-edge AI efficiency research, with a strong ownership mentality and commitment to delivering a presentable result.

Benefits

• A culture that values authenticity and diversity of thought.
• An inclusive and respectable environment with open workspaces and a startup spirit.
• Opportunities to network with industrial pioneers and contribute directly to the digital asset industry.
• Involvement in new projects and system development with personal accountability and autonomy.
• Fast growth, learning opportunities, and attractive welfare benefits, including training and mentoring.