Applied Scientist
Job Summary
As an Applied Scientist on the Connections AI team, you will build and improve machine learning systems that transform complex transaction and communication data into actionable intelligence. You will focus on propensity, ranking, prediction, and decisioning systems, while selectively integrating Generative AI to optimize partner workflows. Your role involves end-to-end model development, from data processing and experimentation to deploying production-ready solutions.
Responsibilities
- Build and optimize applied ML systems for ranking and prediction
- Process messy, real-world behavioral and operational data
- Design and execute rigorous ML/GenAI model evaluation frameworks
- Collaborate with cross-functional teams to deploy production-ready solutions
Required Skills
- 3+ years of professional ML experience
- Strong Python proficiency
- Classical ML (Ranking, Regression, Classification)
- Data evaluation and experimentation expertise
- Experience with large-scale data and unstructured text
Job Details
Key Responsibilities:
- Build and improve applied ML systems, developing models that identify customer readiness and recommend next-best actions.
- Work with large, messy, real-world data including behavioral, conversational, operational, and CRM datasets.
- Design rigorous evaluation frameworks for both traditional ML and GenAI, defining metrics tied to business outcomes.
- Prototype and test new ideas through experimentation, staying current with academic and industry advancements.
- Partner across disciplines with product managers and engineers to translate requirements into measurable objectives.
- Contribute to reliable production science by writing clear, well-tested code for data preparation, training, and integration.
Required Qualifications:
- 3+ years of experience in an academic or industry setting applying machine learning to real-world problems.
- Solid fundamentals in supervised learning, ranking, classification, regression, feature engineering, and model iteration.
- Experience designing and analyzing experiments with an evaluation-first mindset.
- Proficiency in working with imperfect, unstructured datasets and understanding the impact of data quality on model outcomes.
- Strong coding skills in Python and familiarity with common ML tooling.
- Collaborative problem-solving and communication skills.
Preferred Qualifications:
- Experience with NLP, text-heavy products, or communication datasets.
- Experience evaluating LLM-based features (e.g., human evaluation, rubric-based review, LLM-as-judge).
- Experience with marketplace or workflow optimization, including recommendation and decision-support systems.
- Proficiency with big data tooling such as SQL, Spark, or Databricks.
- Advanced degree (M.S. or Ph.D.) in computer science, statistics, machine learning, or a related field.