ML Research Engineer

Join to apply for the ML Research Engineer role at Celestir We need a hands-on ML engineer to implement cutting-edge research in uncertainty quantification and build our open-source confidence calibration toolkit. You'll turn papers into production code that thousands of developers will use. Required Skills Strong PyTorch experience (2 years hands-on) Paper implementation experience - you've implemented at least 3 research papers Python expertise - clean, efficient, documented code GPU optimization - experience with CUDA, mixed precision training Git proficiency - comfortable with collaborative development Primary Responsibilities Implement research papers in PyTorch (focus on confidence calibration methods) Develop and maintain our confidence-kit open-source library Run large-scale experiments on model calibration Optimize inference performance for production use Debug and fix issues in neural network implementations What We Offer Paper co-authorship - build your research profile GitHub fame - your code used by thousands globally Flexible hours - no micromanagement Learning budget - ₹1,500/month or R$100/month Milestone bonuses - up to 20% monthly bonus Conference attendance - virtual passes to NeurIPS, ICML, ICLR Equipment (additional) support after 3 months Seniority level : Entry level Employment type : Full-time Job function : Engineering and Information Technology J-18808-Ljbffr

Location: Manaus, Pernambuco, BR

Posted Date: 11/2/2025