AI Discovery Co-Op - Agentic Personalization

About Ancestry: When you join Ancestry, you join a human-centered company where every person’s story is important. Ancestry®, the global leader in family history, connects everyone with their past so they can discover, preserve, and share their unique family stories. With our unparalleled collection of more than 65 billion records, over 3.5 million subscribers, and over 27 million people in our growing DNA network, customers can discover their family story and gain a new level of understanding about their lives. Over the past 40 years, we’ve built trusted relationships with millions of people who have chosen us as the platform for discovering, preserving, and sharing the most important information about themselves and their families. We are committed to our location flexible work approach, allowing you to choose to work in the nearest office, from your home, or a hybrid of both (subject to location restrictions and roles that are required to be in the office- see the full list of eligible US locations HERE ). We will continue to hire and promote beyond the boundaries of our office locations, to enable broadened possibilities for employee diversity. Together, we work every day to foster a work environment that's inclusive as well as diverse, and where our people can be themselves. Every idea and perspective is valued so that our products and services reflect the global and diverse clients we serve. Ancestry encourages applications from minorities, women, the disabled, protected veterans and all other qualified applicants. Passionate about dedicating your work to enriching people’s lives? Join the curious. Ancestry seeks an exceptional, passionate, and highly motivated Applied AI Science Co-Op to join our team this spring. Our team builds and advances the cutting-edge Agentic AI solutions that drive Ancestry’s personalized discovery, adaptive recommendations, and long-term customer journeys. As an Applied AI Science Co-Op focusing on Agentic Personalization, you will help transition our architectures from traditional embedding-based ranking to LLM-orchestrated recommendation ecosystems. You will research and implement dynamic preference memory, tool-use frameworks (where LLMs control traditional retrieval and ranking models), and simulated evaluation environments. Rather than optimizing for the immediate next click, you will build autonomous pipelines that understand a customer’s evolving research skills and proactively curate long-horizon discovery paths. You will collaborate closely with applied scientists, machine learning engineers, and product partners to translate these advanced multi-agent and reasoning concepts into scalable, real-world AI systems. These efforts are foundational to delivering meaningful, adaptive family history experiences and advancing Ancestry’s leadership in agent-driven AI. This is a part-time, work-study opportunity designed for active graduate students. What You Will Do: Research and develop agentic personalization systems that use customer behavior, genealogy data, and AI to create meaningful, adaptive family history experiences Explore AI memory and personalization architectures , including user profiles, preference representations, event history, and embeddings Develop hybrid agentic recommendation systems where LLM agents orchestrate retrieval, ranking, and personalization models Investigate long-term customer journeys and interactive experiences , moving beyond optimizing for the immediate next click Develop and evaluate agent-based and simulated-user frameworks for testing multi-step personalization and recommendation quality Collaborate with applied scientists and engineers to prototype and deploy scalable AI solutions for personalization, recommendation, discovery, and genealogy Who You Are: Pursuing an advanced degree ( PhD strongly preferred ) in Computer Science, AI, or a related field Experience with applied ML research , including implementing or adapting published ML methodologies; publications in NeurIPS, ICML, ICLR, ACL, KDD, RecSys, or similar venues are a plus Proficient in Python, SQL, and PyTorch , with experience building and evaluating ML systems Hands-on experience with agentic AI , including agent orchestration frameworks such as LangGraph, DSPy, AutoGen, or CrewAI , and concepts such as tool use, planning, memory, and multi-agent systems Experience with LLM inference, fine-tuning, RAG, or retrieval systems ; familiarity with technologies such as vLLM, LoRA/PEFT, Qdrant, FAISS, HNSW, or Two-Tower models is a plus Familiarity with agentic evaluation , including LLM-as-a-Judge, RAG evaluation, or synthetic user simulation, is a plus Additional Information: Ancestry is an Equal Opportunity Employer that makes employment decisions without regard to race, color, religious creed, national origin, ancestry, sex, pregnancy, sexual orientation, gender, gender identity, gender expression, age, mental or physical disability, medical condition, military or veteran status, citizenship, marital status, genetic information, or any other characteristic protected by applicable law. In addition, Ancestry will provide reasonable accommodations for qualified individuals with disabilities. All job offers are contingent on a background check screen that complies with applicable law. For candidates who live in San Francisco, CA, pursuant to the San Francisco Fair Chance Ordinance, Ancestry will consider for employment qualified applicants with arrest and conviction records. Ancestry is not accepting unsolicited assistance from search firms for this employment opportunity. All resumes submitted by search firms to any employee at Ancestry via-email, the Internet or in any form and/or method without a valid written search agreement in place for this position will be deemed the sole property of Ancestry. No fee will be paid in the event the candidate is hired by Ancestry as a result of the referral or through other means.