Эта вакансия в архиве
Посмотреть похожие вакансии ↓3 месяца назад
Research Scientist-Model Efficiency (Intern) (AI)
Описание вакансии
Текст:
TL;DR
Research Scientist-Model Efficiency (Intern) (AI): Making AI models cheaper and faster to serve without sacrificing important quality, with an accent on quantization, sparsity, pruning, speculative decoding, and efficient attention and KV-cache methods. Focus on implementing published techniques, developing new optimizations, and building rigorous evaluations for measurable improvements in throughput and model performance.
Location: Singapore or Austin, US
Company
A technology company developing Bitcoin mining infrastructure, AI cloud capabilities, ASIC hardware, and high-performance computing datacenters.
What you will do
- Implement and adapt published model-efficiency methods on production models and hardware.
- Develop optimization techniques when existing approaches do not meet performance or quality targets.
- Work on quantization, sparsity and pruning, speculative decoding, MTP, attention, and KV-cache optimization.
- Design rigorous evaluations using task-level metrics, controlled comparisons, and honest baselines.
- Own one well-defined project that delivers a presentable result within 3–6 months and present the outcome.
Requirements
- Undergraduate, Master's, or PhD study in Computer Science, Electrical Engineering, Mathematics, or a related field.
- Availability for a 3–6 month internship, with full-time commitment preferred.
- Strong Python programming skills and hands-on familiarity with PyTorch.
- Coursework, self-study, or research experience in model efficiency techniques.
- Strong understanding of transformer internals and rigorous model evaluation.
- At least one substantial project, paper, or open-source contribution that can be explained end to end.
Nice to have
- Formal publication record in model efficiency or related AI research.
- Achievements in academics, engineering projects, open-source contributions, or programming and algorithm competitions.
- Experience with cutting-edge AI efficiency research.
Culture & Benefits
- Inclusive environment that values authenticity and diverse perspectives.
- Startup-oriented atmosphere with open workspaces and opportunities to work on new projects.
- Autonomy, personal accountability, fast growth, and direct impact on AI and digital-asset technologies.
- Training, mentoring, welfare benefits, and professional development opportunities.
- Opportunities to connect with industry pioneers and technology enthusiasts.
Музыка
Осень
Поддержка