• Design, develop, and deploy machine learning and artificial intelligence solutions that drive product innovation and business impact • Work closely with engineering, data, product, and business teams to build scalable models, production-grade ML systems, and intelligent applications • Handle the full machine learning lifecycle, from data preparation and feature engineering to model training, deployment, monitoring, and continuous improvement • Combine strong technical depth with practical problem-solving skills and collaborate effectively with both technical and non-technical stakeholders in a remote environment aligned with U.S. time zones

• Step is looking for an experienced Data Scientist to build and deploy Machine Learning models for all things Risk and Fraud. • You’ll design, develop, and deploy machine learning models to enhance our Risk and Fraud detection systems. • You’ll take the lead in the Risk/Fraud area, providing technical direction and helping shape team’s strategy. • You’ll use SQL to efficiently fetch, transform, and manipulate data. • You’ll write production-grade code to deliver robust machine learning solutions. • You’ll apply statistics to guide experiments, determine appropriate sample sizes, and evaluate model performance. • Partner with Operations Team to quickly respond to rapidly evolving events.

• Integrate open-source and third-party models into our inference platform • Lead fine-tuning initiatives (LoRA, adapters, PEFT, domain adaptation) • Optimise inference workloads for latency, batching, memory efficiency, and throughput • Benchmark model quality vs cost vs performance across modalities • Improve inference startup times and stability under high load • Build evaluation frameworks and internal tooling for model validation • Work closely with Infrastructure and Backend teams on scalable serving systems • Monitor production performance and drive continuous optimisation • Mentor engineers and help raise the ML engineering bar across the team

• Design, develop, and deliver solutions that will enhance and add value to new and existing data assets for broad industry application and aligned with business and market requirements. • Develop algorithmic and rule-based models and analytic capabilities based on rules engine, machine learning, mathematical modeling, pattern recognition, and data mining techniques. • Perform data analysis, feature engineering and model development using a variety of analytic techniques, tools, and programming languages to provide business insights and analytic products. • Create model reports for stakeholder review and model documentation. • Perform pattern recognition model creation and training using various types of algorithms and machine learning modeling techniques for unknown or less well-defined problems. • Develop analytics using Python, PySpark and other big data tools such as AutoML, BigQuery(BQ), BQML, DataPrep and Jupyter Notebook. • Develop analytics solutions using traditional database technology such as SQL and big data technology such as Hadoop, MongoDB, Cassandra, Redis, PostgreSQL and Neo4j, NoSQL databases. • Perform complex data analysis on large volumes of property and consumer data and present model performance to stakeholders.