Applied Science Intern

About this role

Batch: Dec 2027 - 2029. About the role: - Join Ramp's Applied Science team, which develops quantitative models and tools for areas including business underwriting, fraud prevention, spend management, credit, growth and core product experiences. - Interns operate as integrated team members, owning a project end to end while working with engineers, product managers and business stakeholders. - The work involves turning complex business requirements into scalable machine-learning solutions, shipping production code and creating measurable value for Ramp and its customers. What you'll work on: - Manage the full ML lifecycle, covering data exploration, feature engineering, model training, benchmarking, deployment and monitoring. - Apply modern LLMs to new problems and use them to create additional customer-facing capabilities. - Select techniques based on the problem, including deep learning, gradient boosting and causal inference. - Measure results through A/B testing and other statistical experimentation methods. - Partner with product and business teams to turn model outputs and analytical insights into strategy and user-facing functionality. What we're looking for: - Applicants must currently be pursuing a B.S., M.S. or Ph.D. in Data Science, Computer Science, Math, Physics, Economics, Statistics or another quantitative discipline, with expected graduation between Dec 2027 - 2029; graduate-level study is preferred but not mandatory. - Strong foundations in machine learning mathematics, statistics, probability and optimization are required. - Candidates should have strong interest or experience in applying advanced AI, including LLMs and agents, to practical problems. - Python proficiency is expected, including familiarity with tools such as pandas, scikit-learn, NumPy and PyTorch. - Applicants should know SQL and have experience manipulating data in modern warehouses such as Snowflake, BigQuery, Redshift or ClickHouse. - Practical experience preparing datasets and building and evaluating ML models is required. - Strong communication skills are important, including explaining complex ideas to technical and non-technical audiences and using data to construct a clear narrative. - Candidates should be comfortable with ambiguity, move quickly toward shipping solutions and iterate after launch. Nice to have: - Relevant AI/ML publications, projects or previous experience are beneficial. - Knowledge of production-oriented ML engineering practices such as Git, testing and maintainable code is useful. - Experience with orchestration tools including Airflow, Dagster, Prefect or Metaflow is a plus. Compensation and benefits: - The internship pays a monthly rate of $12,500 USD plus a housing stipend and provides an Apple MacBook, catered weekday lunches in the NYC office and a weekly coffee stipend. - Benefits listed for full-time Ramp employees globally include flexible PTO, centralized home-office equipment ordering, a health and wellness stipend, intra-office travel budget and weekly coffee support. - U.S. full-time benefits listed include 100% employee medical, dental and vision coverage with partial dependent coverage, One Medical membership, a matched 401(k), fertility HRA up to $10,000 annually, paid parental leave, pet insurance, office food and drinks, and relocation expense coverage to NYC or SF when needed; Canada benefits include Sun Life health coverage, life/AD&D/disability insurance, fertility drug coverage up to $4,000 lifetime, matched RRSP and DPSP retirement plans, parental leave, an Employee Assistance Program and Lumino Health virtual care; UK benefits include Freedom Elite private medical insurance, eMed x Livi virtual GP and home care, a Penfold workplace pension with salary sacrifice, and parental leave. About Ramp and other information: - Ramp builds finance infrastructure embedded in business spending flows and says it automates more than $200B in annualized spend across 70,000+ companies through functions such as payment authorization, risk detection, spend categorization and book closing; Ramp states that its median customer saves 5% and grows revenue 16% in the first year. - The company emphasizes high agency, urgency, end-to-end ownership and demonstrated building ability over traditional pedigree. - Qualified applicants with arrest or conviction records will be considered in accordance with the San Francisco Fair Chance Ordinance. - Ramp warns candidates that legitimate recruiting communication comes only from official @Ramp.com addresses and that it will not request payment or sensitive personal information during hiring. Note: If you are being referred, contact your referrer so they can submit the application on your behalf.