🛠 Tech Stack

💼 About This Role

You'll work on every stage of LLM development—pretraining, post-training, and serving—for trading workflows at a leading quantitative trading firm. Your models will be deployed firm-wide to accelerate research, engineering, and trading. You'll have access to state-of-the-art research clusters with high GPU-to-researcher ratios and minimal bureaucracy.

🎯 What You'll Do

• Train tokenizers and curate datasets for pretraining
• Design and run pretraining and post-training experiments
• Build evaluation datasets and analyze model performance
• Optimize inference and integrate models into trading systems

📋 Requirements

• Two+ years hands-on experience training or evaluating LLMs at scale
• Direct experience with large-scale pretraining or post-training (SFT, RL)
• Ability to design ablations and update beliefs from evidence
• Strong engineering skills in GPU kernel development or PyTorch internals

✨ Nice to Have

• Published work on a frontier-style LLM result
• Experience with distributed training/inference systems

🎁 Benefits & Perks

• 💰 Discretionary performance-based bonuses
• 🏖️ Competitive benefits package
• 🚀 State-of-the-art research clusters
• 🧠 Supportive engineering and systems teams

📨 Hiring Process

Estimated timeline: 2-4 weeks · AI estimate

1. 1Recruiter Screen· 30 min
2. 2Technical Interview· 60 min
3. 3On-site (Research Talk + Coding)· 3-4 hours