Last week, we released Coveo for Sitecore 4.1, and with it the Coveo for Sitecore Hive framework. Obviously, the documentation followed as well as a high level blog post, but I wanted to explain in my own words what this release is all about.
In my last post, I shared some of the wisdom I gathered over the 4 years I’ve worked with AWS Redshift. Since I’m not one for long blog posts, I decided to keep some for a second post. Here goes!
Organizations are becoming more and more aware of the importance of offering a good search experience on their website. From an increase in purchases on ecommerce websites to getting fewer cases on their support and community websites, companies are realizing that helping users find what they want faster and more easily results in a greater overall user experience.
In this series of blog posts, I will explain why having a good search experience is vital for any website. This blog post will tell you about the importance of the search box.
Over the last 4 years, I have been part of the team that builds the Usage Analytics solution here at Coveo. This solution is based on AWS Redshift, a petabyte scale columnar store. We were early adopters of this data warehousing solution and while it is an awesome product today, I probably don’t need to tell you that we hit some bumps along the way. Here are some of the tips, tricks, and overall best practices we gathered during those years.
Recently, I had to work on an interesting use case where my client wanted to display teasers of premium content to anonymous users in Sitecore. Once the users found what they were looking for, they were either redirected to a login page or to a subscribing page. No secrets, the strategy behind it is to increase conversion rates.
After working on the project, I realized that the business needs behind their request was fairly common. People want to show partial items to anonymous users, but still want them to be relevant and easy to find. Let me introduce you the
Kubernetes: one hip word we see everywhere in the Cloud developer and Devops world. With reasons: Kubernetes does solve problems (and creates others) and simplify a lot of things. In this post we’ll explore how we deployed k8s to production automatically with help from Terraform, Jenkins, and Kops.
At Coveo, we decided that Kubernetes was the tool of choice to run our docker containers in production. This is replacing a “homemade” setup with AWS Opsworks and dockers. It will save money and resources, and enable faster deployments.
After reading the title of this post, you were probably wondering the same thing everyone asked us while working on this project:
And you would be right. Why would we want to remove Coveo’s most useful features: Machine Learning, sorting, and automatically-tuned relevancy?
There are 4 reasons:
- For fun
- To try to get different results each time you execute a query
- To test what we could achieve with the current infrastructure and tools
- But yeah, mostly for fun
This post will cover the road we had to walk to achieve such results.
In my life, I really love to automate recurring tasks. In my private life, I’ve been conceptualizing, building, and developing my home automation for many years. At work, I automate everything: scripts, servers, deployments, etc. And now in the cloud, I am automating the infrastructure.
It is not a new thing; automation is old, but it’s not in all data centers or cloud infrastructures. Let’s see why you would have to automate your infrastructure, except because it’s cool. There are other points to consider.
In my last blog post, I talked about how we use webpack at Coveo, to improve our development process.
One of the most powerful feature offered by it is code splitting.
This is the first blog post of a new series entitled “Build it with Coveo”. The series will present innovative use cases for the Coveo Platform, always including full code samples.
Coveo customers use the platform in a multitude of ways. Many in the consulting and resource management business often ask us if our search technology could help them better match resources and projects.
What if you want to find the best peer/employee with knowledge around “Artificial Intelligence.”? What if you need to add constraints such as “available for the next two months.”?