<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=2634489&amp;fmt=gif">

Infrastructure, Industry Solutions, Interview, Retail, Databases

Building for Scalability, Block by Block: An Interview with Carrefour Links CTO Mehdi Labassi

By Stephen C2C | October 14, 2022

When I ask Mehdi Labassi ( @Mehdi_Labassi ), CTO of Carrefour Links, what he does outside of work, the first thing he mentions is his family. Mehdi spends a lot of his free time playing with his kids. Sometimes they play video games on Nintendo Switch, but they also enjoy hands-on activities like building with Legos. Lego is a popular interest among tech practitioners building products on Google Cloud - - after all, the four letters in "Lego" can also be used to spell "Google." This connection turns out to be a fitting point of departure for an examination of Mehdi's journey to a decision-making role on the technical team at Carrefour Links.

Mehdi began his career as a software engineer, working first in air travel and then moving on to Orange, "the one major telco in France." At Orange, Mehdi led the company's Google Cloud skills center and took part in a major migration to Google Cloud from a historically on-premises infrastructure. "We had a really strong on-prem culture, so we had our own data centers, our own Hadoop clusters with thousands of machines, and the shift to cloud-based services was not something natural," he says. "There was a lot of resistance, and we needed to really show that this gives us something."

Proving the value of the Cloud to a historically on-prem organization required zeroing in on a specific technical limitation of the existing infrastructure: "As I was driving the big data platforms and the recommendations, I do remember we had a lot of issues in terms of scalability." Google Cloud turned out to be the perfect solution to this problem. "Then we tried the Cloud, and we found that instant scalability," says Mehdi. "That's another level compared to what we had on prem, so this is really the proof by experimentation."

 

"You assemble and program the thing, and then you need to understand how each brick works."

 

When Carrefour introduced Carrefour Links, its cloud-hosted retail media and performance platform, in Spring of 2021, Mehdi was immediately interested in getting involved. He reached out directly to the executive team and joined as CTO three months after the company declared the platform. "I joined when the thing just got in production, the first version, the V1. That was kind of a proof of concept," Mehdi says. In the time since - - only a little over a year - - the venture has grown considerably: "We have a lot more data from different verticals, everything that's related to transactions, to the supply chain ecosystem, to finance, a lot more insights, and we are exploring machine learning, AI use cases - so we are scaling even in terms of use cases."

Even a fast-growing platform run on Google Cloud, however, will encounter challenges as it continues to scale. "The first thing is the ability to scale while keeping FinOps under control," Mehdi says. As he sees it, this is a matter of "internal optimization," something he believes Carrefour Links handles particularly well. "The second thing is how to provide what I call a premium data experience for our customers, because we are dealing with petabyte-scale pipelines on a daily basis, and however the end user connects to our data solutions, we want him to have instantaneous insights," he adds. "We leverage some assets and technologies that are provided by Google Cloud to do this."

These are challenges any technical professional managing products or resources on the cloud is likely to face. Overcoming these challenges is also what makes new solutions on the cloud possible. What competencies do IT professionals need to be able to overcome these challenges and pursue these solutions? According to Mehdi, "a good engineer working on the cloud, with this plethora of tools, he needs to be good at Lego." Mindstorms, Lego's line of programmable robots, he explains, require a lot of the same skills to build as machines and systems hosted on the cloud. "You assemble and program the thing, and then you need to understand how each brick works," he says. "I really find a lot of similarities between these activities and what we are doing in our day job."

 

Extra Credit:

 

 


Recent Articles

Data Analytics

Generative AI: Are You Behind?!

Review the latest insights from the AI Readiness Report.
By Bruno Aziza
Industry Solutions

Make "Gen AI Work": Landscape, SLMs vs. LLMs, Cost & More...

Discover the 5 metrics you need to know in order to be an exceptional CEO and Operator.
By Bruno Aziza
Google Cloud Strategy

AI Cheat Sheet

AI is no more and no less the drive to create robots with human minds so they can do everything we do and more. Use this cheat sheet to help decode the space.
By Leah Zitter