Analytics Engineer
Greenspoon is an online supermarket in Kenya. We deliver groceries to customers through express and scheduled deliveries. We are building East Africa’s largest online supermarket, with service and quality that customers can rely on.
We care about our customers, suppliers, employees and the environment. Our values are honesty, quality and impact. As we grow, we need people who take responsibility, think clearly and make the work easier for the next person.
Role
You will help build the next stage of our data platform. Working with analysts, data scientists and business teams, you will help make sure everyone at Greenspoon works from the same definitions and trusted data, and can get answers quickly through dashboards, tools and AI assistants.
You will build and maintain data models, apply engineering practices to our pipelines and follow the team’s standards for quality and documentation. This requires understanding both how our data is built and how the business uses it.
Purpose
Five years ago we began digitizing our operations. A year ago, we started building our data platform. Pipelines now bring data from multiple systems into BigQuery, and Power BI, Looker and Apps Script tools give teams visibility and drive action. We have also started our data science and machine learning work.
Our data helps teams make decisions to optimize deliveries, warehouse operations, commercial performance, finance, and customer satisfaction.
As more teams rely on data, the same metric gets calculated in different ways, pipelines become harder to change and costs grow. Our main objective now is scalability and data trust, and this role will play a direct part in it. We need a platform where a number means the same thing in every report, where errors are caught before the business sees them, and where new use cases can be added without rebuilding what already exists.
Responsibilities
- Data models. Build and maintain dimensional models and semantic layer components for areas such as orders, deliveries, inventory, products, customers, suppliers and finance. Implement core metrics according to agreed definitions so that dashboards, tools and AI interfaces use them consistently.
- Pipelines. Develop transformations in Dataform or dbt. Work with version control, code review, testing and CI/CD so changes are traceable, maintainable and safe to release. Build and maintain data marts for business areas.
- Data quality. Write automated tests and monitoring for freshness, completeness and business rules. Raise issues with Tech when source system changes affect reporting. Investigate discrepancies and help fix their causes.
- Documentation. Document the tables, columns and metric definitions you build, and keep lineage and the data catalogue up to date so analysts and business users can rely on them.
- Governance and security. Apply access controls, data ownership rules and the agreed process for changing metric definitions.
- Cost. Write efficient queries and models, and flag pipelines or dashboards that drive up data warehouse or BI costs.
- Analytics support. Work with analysts and data scientists to build dashboards and analytics solutions for operations, commercial, finance and sales. Help check models, queries and dashboards for accuracy and consistent use of metrics.
- AI integration. Support the integration of our data platform with AI tools, particularly Claude, so people across Greenspoon can access data and insights directly. Help provide the clear definitions and context these tools need to answer accurately.
- Roadmap. Contribute to the data platform roadmap by sharing what you learn from the business and from the data, and help make sure new work is built with reporting in mind.
Approach
- Understand the business. Learn how orders, stock, deliveries and money move through Greenspoon before you model them. Speak with the people who use the numbers.
- Define before you build. Confirm what a metric means, who owns it and how it will be used before it reaches a dashboard.
- Build for the next person. Write models, tests and documentation that someone else can understand, change and trust.
- Follow clear standards. Apply the team’s naming, modelling, testing and dashboard conventions consistently, and suggest improvements where you see them.
- Work across teams. Work with Tech on source systems, with analysts on reporting and with business leads on the decisions the data must support.
- Learn and share. Seek out code review and feedback, and share what you learn with analysts and data scientists.
- Make sensible trade-offs between speed, quality and cost. Be open about risks and ask for input when unsure.
- Close the loop. Check that your changes improve accuracy, reliability and cost, and that they keep working over time.
Requirements
- Experience building data models or pipelines in production, in an analytics engineering, data engineering or analytics role.
- Strong SQL and a good understanding of dimensional modelling. You can build facts and dimensions that give the business consistent metrics.
- Hands-on experience with dbt or Dataform, including testing and documentation, on a cloud warehouse such as BigQuery or Snowflake.
- Familiarity with data quality practices such as automated testing, documentation and access control.
- Working knowledge of at least one major BI tool such as Power BI, Looker or Tableau.
- Comfort with version control (Git) and code review.
- Clear communication. You can explain your work to business leads, analysts and engineers.
- Interest in AI and LLM workflows, and how to give AI tools reliable access to governed data.
- Low ego and curiosity about the business. You accept feedback, own mistakes and connect data work to customer experience and profitability.
- Experience in retail, e-commerce, logistics or another operations-heavy business is a plus.
Expectations
- Judge your work by whether the business trusts and uses the data. Building pipelines is not enough.
- Face the facts. When a number is wrong, say so early, find the cause and help put it right.
- Reduce the margin for error. Use tests and monitoring to prevent mistakes or flag them early.
- Understand the platform. Follow problems from the source system to the dashboard and help resolve them.
- Prioritize and finish. Focus on the problems that matter most, avoid duplicated logic and verify that fixes last.
Success
The models and pipelines you own are tested, documented and use agreed metric definitions. Data issues in your areas are caught before business users find them, and your work runs reliably and efficiently.
Analysts, data scientists and business teams can rely on the data you build, including through Claude or similar AI assistants. You can show what you built, what improved and why it lasted.