Skip to content 
Educational analytics platform
Innive
We built the advanced analytics behind Innive’s K12 360⁰ Remote Learning and Classroom Analytics Platforms that provides insights into the tech needed for different learning models.
Innive is a US-based edtech organisation that contributes to missions of students being able to access better education. Amidst the Covid pandemic, the client set out to help schools to migrate their classes from real life to online, an immensely important challenge, and a technically very difficult one.
Innive launched their K12 360⁰ Remote Learning and Classroom Analytics to help teachers, schools and districts to understand the technology needed for their different learning models. The solution allows stakeholders to evaluate what is available and required, measure usage and assess effectiveness of the programme by measuring student outcomes.
What we did
The Challenge
Schools had to rapidly adjust to life online. However, managing the vast amounts of data that make up a holistic education for a child was no easy task.
There were a myriad of data sources but what is available to the schools will be determined by their resources. All of the data had to be integrated and this included data related to the use of Google Classroom or Zoom, to classes offered, to teachers, students, lessons, dates of terms, syllabus, all matters and needs integrating.
BI tools at the front end would become essential to governments, school districts, teachers and individual students. Ultimately, they could and had to become automated and personalised to each user account, thereby giving confidence to the end user.
integrated learning analytics platform
Working closely with Innive, we created a blueprint for an integrated learning platform to be rolled out to school districts in the US. Our solution enabled comparison across students, teachers, schools and districts of how their remote learning programmes were going. Even if different schools used different technologies to deliver the teaching, the solution supported platform-agnostic engagement and performance metrics.
DevOps
In order to access the school district data, we had to go through a virtual private network (VPN) and then Remote Desktop Protocol (RDP) to a Windows Virtual Machine. This option was just neither scalable, nor efficient. Airflow on Cloud Composer did not work – the limitations were not compatible with what we needed in terms of python packages and dependencies.
We did a direct VPN connection to Google Cloud using a virtual private cloud (VPC). We had tried other solutions such as using Google Cloud VPN, but they would not do the job due to not being able to mask the traffic correctly. Therefore we configured our own instance from scratch.
We built our own Airflow on Kubernetes engine and configured it all from scratch with Terraform and Helm. We then automated the deployment to AirFlow with GitLab CI/CD. With a single command, one could spin up the whole infrastructure including the VPN.
Data engineering
Ingesting data from multiple different education technologies, we had to ensure that in most cases they were all in sync with the master education system. However, there was the massive limitation of having to ingest the data via an API that is not built for large data exchange.
We created an optimised multi-platform flow, which ingested data into BigQuery, allowing our database folks to work on the cool stuff and transform the data into a form that is useful to our clients. We also set up flexible synchronisation scripts, allowing us to push up-to-date data from the master system to other platforms.
Data modelling
We always look for ways to automate and standardise our approach to data warehousing and modelling, using a combination of internally developed tools and open-source technology adapted for our purposes.
The benefits of this approach are twofold: applying best practices and allowing enough time to solve the most challenging questions. In this case, the tool that made the difference was dbtvault, which we used for creating automated data vault processes. As dbtault’s official BigQuery support was under development at the time of the project, we ported it. dbtvault’s code gave us ideas of how to automate a data lake, which we generated without the dbtvault package.
The technology we useD
Understanding technological needs in education
Whilst the pandemic has devastated the learning opportunities for so many young people, this project has helped enable millions of children to have access to high quality education. It has given teachers and school districts the ability to learn more about each child, and curate their learning experiences accordingly.
Powered by advanced analytics, Innive’s K12 360⁰ Remote Learning and Classroom Analytics would enable teachers, schools and districts to understand the technology needed for their different learning models. Teachers could now measure and monitor student engagement and districts could track trends over time, empowering all stakeholders to make informed decisions and plan for unprecedented school years.
The COVID-19 pandemic is overwhelming the functioning of education systems worldwide. School districts in the US are now increasingly seeing the use of technology in education as a lifeline. How do you ensure that students have access to high-quality and equitable education? How can schools plan for the upcoming academic year and deliver quality instruction?
One answer is to analyze data from new and existing sources. To do that, they have to think about the potential of data and analytics for informed planning and decision-making.
Gautham Sampath ITIL, CEO, Innive Inc.
Let’s walk the walk together
Our data expertise translates into cutting-edge, modular solutions that empower people and help organisations thrive in the era of AI.
see Related Stories
We work with modern organisations to strategise and build cutting-edge solutions, help them adopt data & AI innovation, and nurture key competencies for scaling in the long run.
We work with modern organisations to strategise and build cutting-edge solutions, help them adopt data & AI innovation, and nurture key competencies for scaling in the long run.