Want to know if this job is worth applying to?
Nairobi, Kenya
Onsite
Kitsilano Technologies is a leading technology consulting firm based in Kenya, focused on helping organizations accelerate digital transformation through cloud, data, and modern IT solutions.
Kitsilano Technologies designs and delivers data platforms, analytics, and reporting for banks, insurers, manufacturers, and hospitality groups across East Africa. We partner with global technology providers such as AWS and Google Cloud to deliver scalable, secure, and cost-effective solutions that drive real business impact.
We are hiring a data engineer to design the models behind client reporting and to present the results to the client directly. Source systems are rarely clean, and requirements rarely arrive complete.
Design and build the ingestion, transformation, and modelling layers that client reporting depends on.
Own data models end to end: grain, keys, relationships, history, and testing, with the reasoning documented.
Profile, clean, and reconcile source data before it reaches the reporting layer.
Build the analytics layer that answers the client's questions, and interpret what the results indicate for their business.
Present designs, findings, and recommendations directly to client stakeholders, including non-technical audiences, and contribute to scoping at the pre-sales stage.
Excellent communication and presentation skills
Self-Starter – Motivated – Customer-focused
Minimum 3 years building and maintaining production data pipelines that other teams relied on.
Undergraduate coursework, university and personal projects do not count toward this.
Certification required: a current cloud certification within the data, analytics, or machine learning track.
Certifications outside that track do not satisfy this requirement.
Snowflake SnowPro Core, production Snowflake experience, and machine learning deployed to production are each an advantage.
Advanced SQL: window functions, CTEs, set logic and aggregation at the correct grain, with the ability to read a query plan and explain a performance problem.
Relational database and modelling depth: normalisation, grain, key relationships, slowly changing dimensions, and judgement on when to denormalise.
dbt in a production setting: incremental models, snapshots, tests, and a project structure you designed rather than inherited.
Python for data work: analysis, automation, and acquiring data from databases, APIs, and files, including scraping where no interface is provided.
Data cleaning as an engineering discipline rather than a one-off task: profiling a source, reconciling it against a system of record, and holding it to that standard as it changes.
Quantitative and logical reasoning: statistics sufficient to validate a result, and the discipline to justify a design decision under questioning.
Presentation of data: turning a result into a clear chart, a written finding, and a recommendation that a non-technical audience can act on.
Where AI tools form part of how you work, you are expected to explain and defend every line of output you deliver. We regard them as an accelerator, not a substitute for understanding.
The databases, platforms, and tools you worked on directly, the scale of the data, and your own contribution rather than the team's.
At least one data model you designed: its grain, why you selected that grain, and what you deliberately excluded.
One instance where source data arrived incomplete, duplicated, or inconsistent, and what you did to make it reliable enough to report on.
Analytics or reporting you delivered, the question it answered, and how the business used it.
The hardest query or performance problem you have solved, rather than a stated proficiency level.
Certifications held, with the awarding body and current status.
The opportunity to work with the leading global OEMs and their technologies.
Ongoing professional development, with certifications supported and funded.
Career growth within a rapidly expanding technology consulting firm.
