#MeetOurTeam – Thibaud

What does a typical workday look like for you?

It's hard to pinpoint a typical workday as it varies depending on the project stage. However, I usually start by preparing my to-do list or having a stand-up meeting before diving into the day's tasks. This may involve researching solutions online, tackling issues, troubleshooting, and restarting until things are resolved.

Which technologies do you work with?

I am focusing on Microsoft Azure with Microsoft Synapse and now Microsoft Fabric. I use mainly Python and SQL. 

At this years DigX and EAGE you will present the talk - A source of truth approach. Tell us a little bit about this?

This year I will present the talk - A Source of Truth approach - both at DigX in Oslo and at EAGE Digital in Paris - at EAGE Digital I will go together with a Cegal colleague from Stavanger
 
The presentation - A Source of Truth approach – is based on a project I do for one of our customers in oil and gas with three other colleagues from Cegal.
Like many other companies in the field, the customer is struggling with the amount of information and the number of data sources they are dealing with for wellbores data. A lot of the information quality control is done manually and this is a tedious job for the geologists and data managers. 
 
The main goal of the project is to automate this process by integrating automatically the different sources my customer is using and displaying the mismatches between the data sources on a dashboard. 
 
We decided to tackle this task with Microsoft Fabric which is a Software as a Service from Microsoft where we can build end-to-end data solutions. I like it because there is almost nothing to configure before being able to use it and I can do a lot without depending on someone else. I am responsible for performing all the data ingestion and transformation while my Power BI expert colleague is building the dashboard. 
 
One of the challenges in this project is that for one of the sources - Diskos - a lot of the information is located in PDF files and extracting data from them (especially tables) is a difficult task. We will leverage the capabilities of the Large Language Models (LLMs) - such as ChatGPT - to extract data from them. 
 

Read about How AI and language models can change the energy industry here > 

What is important for you and Cegal to achieve success?

Having clear objectives and effective project managers is not something that everyone is fortunate enough to have, but being surrounded by supportive and giving individuals is key to success.

Do you use AI in your job? How?

I use Copilot GitHub to help me fix bugs and code faster and as part of an actual project I will use LLMs to extract data from unstructured text

What drives you at work and in your role, what's fun?

I like to learn and build so I get the opportunity to learn with every new assignment.  

The fun is to get challenged by the customer and to solve it technically within the constraints of the technology stack we build the solution with. 

What is the best thing about working in Cegal?

Our manager gives me the freedom and flexibility on how to work as long I deliver. 

Read more about what it's like working at Cegal >

Favorite Podcast?

Super Data Science Podcast – It is a very old podcast (750 episodes) that helps me to stay up to date in the field of data science. 

What is your favorite quote?

Practice makes the master.

Is there anything else fun we should know about you?

After spending a decade in the modeling industry and previously designing new medicines with computers, my life took a turn when I had my aha moment while using Python for the first time five years ago. This pivotal moment led me to transition into the exciting field of data science. Recently, I had another enlightening experience when I discovered the innovative world of low-code/no-code technology.

Additionally, I am proud to be the founder of Pint of Science Norway, where we curate an annual three-day festival. During this event, we transform bars into venues for scientists to share their research with a curious audience.