Principal AI Engineer, Rula

• 10+ years of software engineering, including 7+ years of experience designing and scaling distributed systems, and 5+ years building and deploying production-grade ML applications
• 5+ years of deep, hands-on experience building and optimizing search, ranking, relevance, or recommendation engines at scale (e.g., Learning-to-Rank, collaborative filtering, deep recommender systems, semantic vector search)
• 5+ years of strong programming experience in Python and at least one backend language (preferably TypeScript, Java, or Go), combined with 2+ years of experience building AI-powered products using foundation models (OpenAI, Anthropic, Gemini) and LLM integration patterns (RAG, agents, etc.)
• Strong proven track record working with both traditional search/retrieval infrastructure (e.g., Elasticsearch, OpenSearch) and modern vector databases (e.g., Pinecone, Weaviate, FAISS, Milvus)
• 3+ years of experience with MLOps, data pipelines, and rigorous evaluation systems (offline metrics like NDCG/MAP, online A/B testing), with a demonstrated ability to define technical strategy and guide the broader AI/ML infrastructure ecosystem

Desirable

• At Rula, this role owns the applied AI and ML solutions that determine what, how, and how well AI shows up across mental healthcare
• The focus isn’t on isolated prototypes, but on building production-grade AI systems that directly power patient-provider matching, clinical workflows, and patient engagement
• While your immediate focus will be driving search ranking, relevance, and patient-provider matching, your broader scope encompasses our overarching personalization and recommendation strategies, alongside directing the foundational ML platforms that support them at scale
• You’ll operate where applied research, system architecture, and real-world clinical constraints meet
• The work spans developing advanced recommendation engines and retrieval models, applying NLP and generative AI to clinical workflows, setting technical standards for AI safety, and making principal decisions about how machine learning is introduced into a high-stakes environment
• The goal is simple but hard: leverage applied AI—from intelligent matching and deep personalization to broader AI solutions—to make care more accessible and effective without compromising trust or outcomes
• This is a deeply hands-on role with real technical ownership
• You’ll develop and fine-tune models (e.g., Learning-to-Rank, embeddings, recommendation algorithms, and LLMs), design both product-facing AI architectures and core ML infrastructure, unblock complex engineering problems, and act as a technical multiplier for engineers working across product surfaces
• Success here means shipping applied AI features that tangibly improve match quality and the user experience, while simultaneously building scalable, durable AI foundations that others can build on confidently
• The opportunity is less about chasing the AI frontier for its own sake and more about shaping how applied ML responsibly becomes part of everyday mental healthcare
• You will set the direction not just for our immediate relevance and recommendation systems, but for how the entire AI ecosystem—from infrastructure to advanced clinical applications—evolves to actively improve patient outcomes over time