← Projects
Software2025

Kalkyle – Slaughter Analytics Platform

Backend platform for analyzing carcass yields, cutting patterns, and production efficiency in the Norwegian meat industry.

Role

Backend Developer & Data Engineer

Organization

Animalia

Used by

Animalia and Nortura

Technologies

Python, Flask, SQL Server, Parquet, Docker

Kalkyle analytics dashboard

The Problem

Meat processors generate large amounts of production data, but analyzing how carcasses are transformed into products can be challenging. Different slaughterhouses use different cutting patterns, products, and production processes, making it difficult to compare performance and identify opportunities for improvement. Animalia wanted a platform that could automatically process slaughter data and provide detailed analytics on yields, weights, and processing efficiency.

My Contribution

I designed and implemented the backend platform responsible for data ingestion, business logic, caching, and API delivery. The system automatically collects production data, transforms it into a hierarchical product structure, and exposes analytics through a REST API used by the frontend dashboard.

Product hierarchy model

Product Hierarchy Modelling

One of the key challenges was modelling how carcasses are broken down into cuts and final products. I developed a configurable hierarchy system that represents relationships between carcasses, cuts, and processed products. This allows users to analyze yields and weights at multiple levels of the production chain.

Statistics and histogram view

Analytics Dashboard

The processed data is presented through interactive dashboards showing:

  • Average yields and weights
  • Product distributions
  • Processing statistics
  • Variation between cutting patterns
  • Exportable reports

The platform enables production specialists to quickly understand how raw material is converted into finished products.

Configurable Rules Engine

To avoid hardcoding slaughterhouse-specific business logic, I developed a JSON-driven rules engine that defines product relationships and cutting patterns. This enables domain experts to modify production definitions without requiring changes to the underlying application code.

Technical Architecture

The backend consists of a modular architecture designed for scalability and maintainability. A JSON-based rules engine separates business logic from application code, allowing domain experts to update cutting pattern definitions without requiring software changes.

SQL Server
Data Processing
JSON Rules Engine
Parquet Cache
REST API
Frontend Dashboard

Impact

The platform is actively used by Animalia and Nortura to analyze production performance and carcass utilization. By automating data processing and centralizing analytics, the system provides production specialists with faster access to key performance indicators and more consistent reporting across slaughterhouses.

Key Contributions

  • Designed and implemented the backend architecture.
  • Built a configurable JSON-driven rules engine.
  • Developed data ingestion and caching pipelines.
  • Implemented analytics APIs used by the frontend dashboard.
  • Collaborated with domain experts to model slaughterhouse business logic.
  • Deployed and maintained a production system used by industry stakeholders.

Stack

PythonFlaskSQL ServerParquetDockerREST APIJSONPandasBackend DevelopmentData Engineering