Research Engineer-Model Efficiency (Intern)
Core
Optimize large language models for cost and speed by implementing techniques like quantization, sparsity, pruning, and speculative decoding to make inference cheaper and faster without sacrificing quality.
Role type
Research Engineer (Model Efficiency) Intern
Builds
Optimized model inference pipelines and evaluation frameworks for Bitdeer's AI cloud and mining solutions.
Domain
Artificial Intelligence / Large Language Model Optimization
Deliverable
production ML models
Required skills
Python, PyTorch, Transformer architecture internals, Model quantization, Sparsity and pruning, Speculative decoding, KV-cache optimization, Rigorous evaluation metrics
Preferred skills
Academic publications in model efficiency, Open-source contributions, Algorithm competition achievements, End-to-end project ownership
Technologies
PyTorch, Python
Responsibilities
Implement and adapt published efficiency methods on models and hardware, Develop custom optimizations where existing methods fall short, Build evaluation discipline to validate performance claims, Own and present a well-defined project producing a real result within 3–6 months
Seniority
Intern