Data Scientist (Remote)

Reimagine access. Reimagine ownership. Build the future with Cinch.

Cinch is Southeast Asia's leading circular tech subscription company, redefining how people and businesses access technology. Instead of adding to the growing e-waste crisis, we enable affordable, sustainable access to devices through subscriptions that maximize reuse and extend their lifecycle.

We operate across B2C, B2B, and B2B2C channels, including major partnerships with global leaders like Samsung. Our model sits at the intersection of powerful mega trends—sustainability, embedded finance, and the shift from ownership to access—creating a massive opportunity to reshape the future of tech consumption.

At Cinch, we are builders with high standards, a bias for action, and a shared mission to create a generational company that outlasts us. Freshly funded by leading institutional investors and growing rapidly, we're creating an environment where ambitious people can do the best work of their careers—and where the work has real, lasting impact.

The Role

We are looking for a Data Scientist to build the credit models that decide who Cinch lends to, on what terms, and how we collect. Every subscription at Cinch is a customer relationship and a credit exposure on a physical device, so your models directly shape growth, losses and asset returns.

You will own credit and risk models end to end: from data exploration and feature design through validation, deployment and monitoring. You will work with SQL and Python, partner with Data, Product and Engineering teams, and turn model outputs into decisions the business can act on. You are expected to be AI-literate and comfortable working with Claude. We expect responsible AI use, and the correctness of all work submitted remains the author's own responsibility.

Responsibilities

Risk and credit modelling

• Build, validate and maintain credit scoring models for new and repeat subscribers - A Score and B Score models.
• Develop early-warning and delinquency models, and support collections strategy and prioritisation
• Run experiments and champion/challenger tests to prove a model improves approval rates, losses or recovery before it scales
• Track model performance over time (stability, drift, bias) and retrain when it degrades
  
  

Data and features

• Collect and blend data from multiple sources (applications, payments, devices, bureau and alternative data) into clean, documented feature sets
• Work with data analysts and engineers on dimensional models and data pipelines so features stay consistent between training and production
• Define and track the risk KPIs the business relies on (approval rate, default rate, loss per cohort, recovery rate)
  
  

Decisioning and partnership

• Turn model outputs into clear decision rules and recommendations to guide credit policy at Cinch.
• Present findings and trade-offs to leadership in plain language, including what the model cannot tell us
• Partner with Risk & Data leadership on credit policy and regulatory expectations in each market
  
  

Standards, AI and mentoring

• Document models so another data scientist can reproduce and audit them
• Use AI tools (Claude Desktop, Claude Code) to speed up exploration, coding and documentation, and set the standard for responsible use with sensitive customer data
  
  

Qualifications

Must-haves

• Bachelor's or Master's degree in Statistics, Mathematics, Computer Science, Data Science or a related field (or equivalent experience)
• 3-6 years of experience in data science or advanced analytics, including building models that went into production
• Strong SQL and Python (including familiarity with all the relevant frameworks and packages such as pandas, scikit-learn, XGBoost or similar) for data manipulation and modelling
• Solid grounding in statistics and machine learning for classification and risk: logistic regression, gradient boosting, calibration, cross-validation, and evaluation metrics such as AUC, KS and Gini
• Experience with model validation, monitoring and documentation, and explaining model results to non-technical stakeholders
• Working knowledge of dimensional modelling and data pipelines, and the ability to work with analysts and engineers on feature data
• Working proficiency with AI tools
  
  

Nice-to-haves

• Experience in fintech, lending, consumer credit, e-commerce or subscription businesses
• Credit scorecard development, collections analytics or fraud detection
• Experience with cloud data warehouses and modern data stack tools (for example BigQuery, dbt, SQLMesh)
• Experience with BI tools such as Apache Superset or Looker
• Experience with workflow automation tools such as n8n, and with deploying models to production
• Familiarity with regulatory expectations for credit data and model use in Indonesia or other Southeast Asian markets