Member of Technical Staff - Machine Learning Research Engineer, Post-Training

What you'll actually do

• Fine-tune both general-purpose and specialized models to align with specific use-case requirements.
• Design post-training datasets and develop rigorous evaluation frameworks to target and quantify model improvements.
• Participate in post-training research discussions and collaborate with cross-functional teams to translate novel ideas into practice.

What makes you stand out

• Hands-on experience in model fine-tuning, particularly with direct alignment and reinforcement learning techniques.
• Expertise in generating, curating, and evaluating high-quality post-training data to effectively target model capabilities and limitations.
• Strong understanding of technical challenges in post-training optimization, including hallucination mitigation, reasoning improvement, and safety guardrail enhancement.
• Proven ability to publish research, models, or novel techniques in post-training optimization.
• Experience in developing both general-purpose and specialized domain models.
• Successful track record of building automated pipelines for continuous model evaluation and improvement.

What you'll get

• Advance the state-of-the-art in model post-training techniques.
• Gain experience working on cutting-edge AI systems that push the boundaries of LFM performance.
• Become an expert in the entire post-training lifecycle from data generation to evaluation.