Senior Architect, Model Performance and Post-Training

What you'll do

• Develop frameworks to measure and analyze coding model performance across various software engineering tasks.
• Optimize post-training pipelines including supervised fine-tuning, preference optimization, and reinforcement learning.
• Analyze benchmark results, training dynamics, and model failure modes to identify improvement opportunities.
• Improve model performance through targeted training interventions, reward design, and curriculum development.
• Build scalable and reproducible infrastructure for training experiments and evaluation pipelines.
• Design experiments to identify performance bottlenecks and understand complex model behaviors.
• Translate model insights into production-ready capabilities for real-world developer productivity.

What we're looking for

• Bachelor's degree or equivalent experience in Computer Science or AI.
• 12+ years of meaningful work experience.
• Excellent programming skills in Python.
• Strong understanding of large language models, transformer architectures, and training/evaluation of generative models.
• Foundational understanding of reinforcement learning algorithms, including MDPs, policies, and reward design.
• Deep understanding of software engineering fundamentals, code generation, automated testing, and program correctness.
• Proven track record leading model development life cycles, including dataset versioning, experiment tracking, and evaluation pipelines.
• Experience training state-of-the-art coding models or models for software engineering agents (preferred).