Kjetil Dyrland
Data engineer in Oslo
I work on data platforms and machine learning. At Tet Digital I build dbt models on Snowflake. Before that I spent a year and a half at Lerøy Seafood moving their data to Databricks on Azure and putting models into production, and two and a half years at Bouvet on data pipelines for Equinor.
I have an M.Sc. in software engineering and machine learning and have co-written two papers on classifier evaluation. Through Dyrland Data & AI I take on data and AI projects, and I build iOS apps.
Experience
-
Sep 2026 – now
Tet Digital AS
Data Engineer
I build dbt models on Snowflake. Raw sources come in, pass through tested staging and intermediate layers, and come out as documented data products that analysts and dashboards rely on.
- Layered dbt models from source to data product
- Tests and documentation on every model
- SQL written for Snowflake, linted and reviewed
Snowflake, dbt, SQL
-
Jan 2026 – now
Dyrland Data & AI
Founder
My own company. Data and AI projects for clients, such as invoice control for Entro, and the iOS apps further down this page.
- Invoice control inside Entro's customer portal
- Eight apps on the App Store
Next.js, TypeScript, Postgres, Swift, Cloudflare
-
Jan 2025 – Aug 2026
Lerøy Seafood
Data Engineer
I led the move of Lerøy's on-premises data infrastructure to Databricks on Azure, and took several machine-learning models from idea to production.
- Price prediction for spot salmon, running in production
- OCR model that reads bills of lading
- An internal AI chatbot
- Pipelines in Delta Live Tables and PySpark, governed through Unity Catalog
Databricks, Azure, PySpark, Delta Live Tables, Unity Catalog
-
Aug 2022 – Dec 2024
Bouvet
IT Consultant
Data engineer on an Equinor project. Large-scale data processing in Databricks on Azure, where I made heavy ETL jobs faster and more reliable.
- Big-data pipelines for Equinor
- ETL performance and reliability work
Databricks, Azure, PySpark
-
Aug 2018 – Jun 2022
Western Norway University of Applied Sciences
Teaching Assistant
Courses in programming, algorithms and data structures, systems development, operating systems, distributed systems and machine learning. I graded assignments and exams and helped students in the lab.
Education
M.Sc. Software Engineering and Machine Learning
HVL and University of Bergen, 2020 to 2022
Grade A. Thesis on machine learning in drug discovery, with the Mohn Medical Imaging and Visualization Centre.
Exchange semester in computer science
University of California, San Diego, autumn 2019
B.Sc. Computer Engineering
Western Norway University of Applied Sciences, 2017 to 2020
Thesis: a mobile app for Kronbar on Android and iOS.
Databricks Certified Data Engineer Associate
Databricks, September 2024
Research
Two papers I co-wrote during my master's, on how to judge and use a classifier when different mistakes cost different amounts.
Does the evaluation stand up to evaluation? A first-principle approach to the evaluation of classifiers
Kjetil Dyrland, Alexander S. Lundervold and P.G.L. Porta Mana. arXiv:2302.12006, February 2023.Why F1, MCC and AUC can pick the wrong classifier, and what to measure instead. 2023Don't guess what's true: choose what's optimal. A probability transducer for machine-learning classifiers
Kjetil Dyrland, Alexander S. Lundervold and P.G.L. Porta Mana. arXiv:2302.10578, February 2023.Turning a classifier's scores into real probabilities, so decisions can account for what mistakes cost.Tools
- Data
- Snowflake, dbt, Databricks, PySpark, Delta Lake, Delta Live Tables, Unity Catalog, PostgreSQL, Supabase
- Machine learning
- Decision theory, Bayesian methods, classifier evaluation, XGBoost, CNNs, MLflow, LLMs
- Cloud
- AWS, Azure, Cloudflare, Vercel, Docker, GitHub Actions
- Languages
- Python, SQL, TypeScript, Swift, Rust, Java
- Apps and web
- SwiftUI, SwiftData, React, Next.js, FastAPI, Xcode Cloud
