Senior individual contributor role building models and data products for the PAYG LPG clean cooking business, judged by commercial results rather than model metrics alone.
Deliver 2-3 data products into commercial or operations use within the first year, each with a measurable business impact.
Build the business unit's approach to pricing and promotion analysis, including elasticity, uplift measurement, and experiment design.
Set up monitoring for shipped models, with retraining triggers and a named owner for each.
Spend time in the field with sales agents and customers to understand their needs.
Turn business problems into statistical questions and present findings to technical and senior audiences.
Prioritise work by its expected effect on activation, refills, retention, or unit economics.
Manage stakeholder expectations, scope, and timelines for your own workstreams.
Design, build, and evaluate ML models for churn, credit and payment behaviour, demand forecasting, anomaly detection, and segmentation.
Apply Bayesian and probabilistic methods to quantify uncertainty in pricing, promotions, and marketing decisions.
Design and analyse experiments, including A/B, geo, and quasi-experimental tests.
Perform exploratory analysis, feature engineering, and data wrangling on large datasets.
Own models from problem framing through deployment-ready handoff, and share monitoring responsibility with data engineering.
Track model performance and recommend retraining or redesign when it degrades.
Write clean, reproducible, well-documented code to team standards.
Requirements
A degree in a quantitative field such as computer science, statistics, maths, engineering, or economics; an advanced degree is a plus.
A track record of data products that measurably moved a business outcome.
Ability to understand customers and estimate the P&L impact of a data product before building it.
Strong communication skills with non-technical commercial leaders.
Experience managing upward on product needs, trade-offs, and timelines.
5-8 years in data science or applied ML, with at least 2 years owning data products end to end.
Strong classical ML skills and the judgment to choose simple methods when they suffice.
Solid Bayesian and probabilistic modelling skills, with experience in PyMC or Stan.
High proficiency in Python and strong SQL, including complex queries and window functions.
Deep knowledge of model evaluation, including cross-validation, calibration, and business-aligned metrics.
Experience with experiment design and hypothesis testing.
Experience with cloud data warehouses and experiment-tracking tools (AWS and MLflow, or equivalents).
Strongly preferred: experience in PAYG, fintech lending, telco, or emerging-market consumer businesses.
Nice to have: survival modelling, causal inference, or marketing mix modelling; operations research or optimisation; MLOps and model deployment on AWS.
Done for you
We'll apply to this for you
Denidel's HR team submits and tracks the application on your behalf.
