Your Impact at LILA

We are growing our Applied AI org and seeking a Senior / Staff Machine Learning Engineer, Applied AI with expertise in model post-training, evaluation, and production-oriented ML systems. You’ll shape how Lila’s AI models are adapted to customer-specific scientific needs, with a focus on turning frontier model capabilities into reliable workflows that can be evaluated, iterated and used in real customer contexts.

Applied AI sits at the intersection of AI Research, model engineering, and product deployment. The team partners closely with AI Researchers and Software teams to adapt Lila models to customer workflows, improve model quality through post training and driving how harness engineering shapes model behavior inside the product.

This is the right role for someone who wants to go deep on model adaptation and evaluation, take real ownership of hard technical problems, and grow into broader scope over time.

What You'll Be Building

• Apply hands-on expertise to close the last-mile gap between Lila's model capabilities and customer-specific scientific workflows.
• Run post-training work using approaches such as SFT and RL (DPO, PPO/GRPO) to align model behavior with customer-specific requirements and feedback.
• Run evaluation loops that measure model quality, reliability, and customer fit.
• Turn customer learnings, data signals, and evaluation results into concrete model improvement cycles.
• Partner with AI researchers to translate model advances into reliable, usable capabilities.
• Work with Software to integrate model behavior into end-to-end product workflows.
• Debug complex model failures using traces, evaluations, customer context, and scientific feedback.

What You'll Need to Succeed

• Minimum 1-2 years of hands-on post-training experience (SFT, RL methods such as DPO/PPO/GRPO) and evaluation system design, gained from having solved these problems many times before.
• Strong software engineering skills in Python and modern ML frameworks like PyTorch.
• A demonstrated ability to debug ambiguous, high-stakes model behavior quickly, using data, traces, logs, and qualitative feedback - the kind of judgment that only comes from having seen many failure modes before.
• Deep familiarity with large language models, multi-modal models, or agentic AI systems.
• Clear communication skills for translating customer needs into technical approaches, and for explaining complex model behavior to both technical and non-technical audiences.

Bonus Points For

• Experience adapting models for customer-facing or production workflows, ideally in scientific, technical, or data-intensive domains.
• Experience with RL post-training, such as RLHF, GRPO, or tool-augmented RL.
• Experience building evaluation harnesses, model monitoring, or quality dashboards.
• Experience training MoE architectures.
• A track record of mentoring engineers and being sought out as a go-to technical expert, rather than a people manager or strategy owner.