The story
Project Background
The Origin
Inspired by real-world DevOps struggles in Swiss enterprises.
The Challenge
Ensuring high-quality software while reducing staging effort
The Vision
AI-driven automation for seamless, efficient software releases.
The Journey
Backed by research, industry partnerships, and cutting-edge AI.
The context
Traditional software staging processes rely heavily on human judgment and manual interventions, particularly at key quality checkpoints. These gates – spanning from continuous integration through test automation and verification to production deployment – serve as key quality control mechanisms.
However, their current implementation often creates a paradox: while designed to ensure software quality and reliability, the manual nature of these gates can introduce delays, inconsistencies, and complexity into the very process they are meant to optimize.
The challenges
Delivering high-quality software rapidly is essential for maintaining a competitive edge, but it comes with significant challenges in the staging process. Quality Gates (QG) verification, essential for ensuring software reliability and performance, is often complex and time-consuming within DevOps pipelines.
Achieving high-velocity delivery demands advanced automation, efficient data analysis, and reduced cognitive load on development and operations teams.
The Vision and Solution
While organizations strive for faster and more reliable software delivery, the traditional approach to quality control creates significant bottlenecks and inefficiencies. AI-SQUARE emerges as an intelligent decision-support platform specifically engineered to advance software staging management in DevOps environments.
By leveraging and combining AI, Machine Learning, and Knowledge Graphs, AI-SQUARE transforms what has traditionally been a human-based, manual decision making process into one that is faster, smarter, and more reliable. AI-SQUARE represents a paradigm shift in how organizations approach quality gates, offering a data-driven, automated approach to what has historically been a time-intensive and complex task.
The initial journey
The journey of AI-SQUARE reflects strategic and collaborative efforts that led to its successful approval by the Swiss Innovation Agency. Starting with a Databooster event and initial brainstorming in 2022, the project evolved through expert consultations, strategic partnerships with academic institutions, and shaping workshops.
The proposal was meticulously crafted and revised with input from scientific teams and Innosuisse mentors. The project was approved in January 2024 and has been in continuous development since.
The Foundations
Swiss Digital Network’s Digital Highway is a comprehensive blueprint for delivering
reliable digital solutions at high velocity, built on five essential engineering pillars:
Effective Site Reliability Engineering (SRE), Continuous Delivery (CD), MLOps,
Observability Engineering, and Continuous Verification.
This foundation inspired AI-SQUARE, which evolved from these principles to
address the challenges of software staging management specifically.
Every software release must go through development, testing, and production stages. Current challenges include:
Too many manual quality assurance steps.
Inconsistent decision-making.
Delays in bringing innovations to market.
AI-SQUARE integrates Knowledge Graphs, Machine Learning, and AI-driven decision-making to automate software staging and enhance DevOps efficiency.
Partners & teams
The AI-SQUARE project represents an innovative collaboration between industry and academic research, funded by Innosuisse. This initiative brings together several key Swiss institutions including the HE-ARC Engineering School, ZHAW School of Engineering, HEPIA Geneva, and Swiss Digital Network.
The project focuses on three main research pillars
Interaction technologies research, emphasizing interoperability and maintainability
Data Science and Computational Intelligence research, focusing on AI/ML applications and hybrid complex systems
Software engineering research, centered on knowledge graphs and continuous integration
The implementation aspects cover product management, testing, DevOps, MLOps, industrial engineering, deployment and releases. The project aims to bridge the gap between cutting-edge research and practical industry applications, with a particular emphasis on knowledge graphs for dynamic software QA and transformation-based analysis for unstructured data.