Content recommendation engine for article suggestions
Recommendation System
Gov
app development
Summary
WKO Inhouse (wko.at), a shared service center for the Austrian Chambers of Commerce, faced significant performance issues and unreliable data transfer workflows. They engaged Postdata, a boutique data science and development company, to optimize API performance, automate long-running tasks, and establish a scalable, maintainable cloud infrastructure.
Within two weeks, Postdata improved API response times from 8 seconds to under 0.9 seconds and introduced robust automation pipelines. By leveraging Terraform for Infrastructure as Code (IaC), GCP Cloud Functions, and Cloud Run Jobs, Postdata ensured an efficient and secure system that continues to evolve with the client’s needs.
"The work is completed quickly and with good quality. Their friendly communication and flexible approach are unique. The team is thorough in their security reviews and focuses on delivering tangible improvements that we can measure and see in our system’s performance." WKO Inhouse
Project overview
WKO Inhouse GmbH is an independent company of the Austrian Chambers of Commerce. They serve as a service provider for all departments of the Austrian Chambers of Commerce, offering a wide range of support including IT solutions, data center services, and more.
Performance Improvement
Challenge: API response times averaged around 10 seconds, causing user frustration.
Solution: Refactored the FastAPI-based application, optimized queries and architecture, and reduced 99th percentile response times from 8 seconds to under 0.9 seconds.
Job Automation
Challenge: Long-running data transfer processes and broken scripts required significant manual intervention.
Solution: Used GCP Cloud Functions and Cloud Run Jobs to schedule and automate tasks such as reindexing data, updating databases, and publishing messages to queues.
Infrastructure as Code (IaC)
Challenge: Infrastructure was scattered and inconsistently deployed, making it hard to maintain.
Solution: Leveraged Terraform for reliable and repeatable deployments across development, testing, and production environments.
Security & Compliance
Conducted a security audit to patch vulnerabilities.
Ensured data was transferred securely between BigQuery, Firestore, and PostgreSQL.
Results
Response times dropped from 8 seconds to 0.7 seconds (99th percentile).
WKO Inhouse noted that Postdata’s speed of delivery and high-quality outputs exceeded expectations.
Communication primarily happened through email and messaging apps, and WKO Inhouse highlighted Postdata’s flexibility and friendly collaboration as major strengths.
Project duration:
1 year and ongoing
Team
4
2 Software Engineers, 1 DevOps Engineer, 1 Data Scientist
Technologies
Core Technologies: Python & FastAPI: For building high-performance APIs. GCP (Cloud Functions, Cloud Run, BigQuery, Firestore): Streamlined serverless computing, data storage, and analytics. Terraform: Automated the setup and teardown of cloud resources, enabling Infrastructure as Code (IaC). PostgreSQL: A robust relational database for structured data. Monitoring & Alerting: GCP Monitoring with custom alerts for immediate issue detection. CI/CD: Cloud Build to automate testing and deployments.
Tech challenge
WKO Inhouse hired Postdata to improve a recommendation system, provide custom software development, cloud consulting, AI consulting, and API development. The primary goal was to reduce API latency and secure the system.
Solution
Postdata built a content recommendation engine for article suggestions and optimized APIs. They conducted a full security audit and leveraged Terraform to deploy and manage cloud infrastructure, resulting in a consistent and efficient process for new deployments.
