top of page
INTERNSHIPS
YOU CAN SEE MORE INFORMATION ABOUT MY INTERNSHIPS BY CHECKING MY LINKEDIN ACCOUNT:
EMBRAER
CONTEXT
This internship was part of Embraer’s winter internship program, an initiative designed to attract students studying abroad and expose them to real industrial projects within the company. Embraer is the third largest aerospace manufacturer in the world, with a strong presence in commercial, executive, and military aviation.
I was allocated to the MISCLAW team (Mission Control Laws) of the KC-390 program, a multi-mission military transport aircraft. The team is responsible for mission-related control laws and their interaction with aircraft systems, operating in a highly multidisciplinary environment that involves avionics, systems engineering, and simulation.
WHAT I DID
Military aircraft such as the KC-390 are delivered to different countries, each with specific operational requirements and mission profiles. As a result, systems are frequently adapted, updated, or newly developed. Whenever such changes occur, the Onboard Maintenance System (OMS) must be updated. The OMS is responsible for displaying system states and parameters to the pilot in real time, as well as generating maintenance reports for ground crews.
Before my project, the design of OMS pages followed a manual and inefficient workflow. Engineers would model the intended cockpit display pages using PowerPoint slides, which were then sent to another team responsible for implementing the pages directly in the aircraft avionics. This process often led to rework, interpretation errors, and duplication of effort, especially when system parameters needed to be modified or reused in simulation models.
My project consisted of developing a new interface inside an internal Embraer software tool that allowed OMS pages to be created directly within the engineering environment. The goal was to replace static visual mockups with a structured, editable model that could be reused downstream. The interface had to be as fast and intuitive as PowerPoint, even though the software was not originally designed for this purpose.
The solution was built using a combination of native software blocks and custom components developed in C++. The result was a simple, modular, and easily editable page model that significantly reduced rework for the avionics implementation team. In addition, the new approach enabled a direct link between the OMS page definition and the team’s Simulink mission models, allowing system parameters to be reused consistently across simulation and avionics design.
The project was successfully integrated into the team’s workflow and validated within the KC-390 mission modeling environment.
Images from this project cannot be shared due to confidentiality restrictions.
LEARNINGS
Beyond the obvious technical development in C++ and system modeling, this internship significantly strengthened my engineering mindset. I learned to approach problems with critical thinking, questioning what was actually being requested, understanding constraints, and identifying the most effective and sustainable way to deliver a solution. This shift from simply “implementing tasks” to thinking in terms of efficiency, impact, and integration was one of the most valuable outcomes of the experience.
More broadly, the internship clarified why aerospace engineering resonates so strongly with me. Not because of a generic fascination with airplanes, but because an aircraft represents one of the highest levels of human knowledge integration. It brings together mechanics, electronics, software, control theory, human factors, certification, and operational constraints into a single coherent system. Experiencing this complexity firsthand made it clear that aerospace engineering is a concrete expression of both my personal interests and my professional profile.
I also learned that no meaningful engineering work is done in isolation. Collaboration, communication, and teamwork are essential, especially in large and complex organizations. Progress depends on the ability to interact with different teams, understand their constraints, and build solutions that fit into a much larger system than one’s own scope.
Finally, this experience taught me a fundamental lesson about engineering practice: strong technical foundations are necessary, but they are rarely sufficient on their own. You are not always fully prepared for the problems that arise. What truly matters is the ability to learn quickly, adapt, be creative, and stay motivated. With curiosity, initiative, and genuine interest in what you are building, even unfamiliar challenges become solvable.
For the invaluable learnings I gained from this experience, I am deeply grateful to everyone involved, and especially to Briza Aiki, Karen Veloso, and my mentor Mateus Marocolo. You added tremendous value to my life. Thank you.
For confidentiality reasons, I'm not allowed to expose the results and the developpement of my work
safran
CONTEXT
During my internship at Safran Tech, I worked on a Research & Development project focused on applying Machine Learning and Artificial Intelligence to industrial thermochemical treatment data.
Nitriding is a surface-hardening process used to improve the mechanical properties of critical components, particularly in aerospace applications. However, characterizing these treatments through measurements such as hardness, residual stress, and nitrogen concentration profiles can be expensive, time-consuming, and sometimes destructive.
The project's objective was to leverage existing industrial datasets to develop predictive and generative AI models capable of recovering missing information and estimating material properties from process parameters or other available measurements. By reducing the need for costly experimental characterization, these approaches aim to accelerate R&D and support more efficient industrial decision-making.
WHAT I DID
-
Generative AI: Implemented a Conditional Variational Autoencoder (CVAE) for probabilistic missing data imputation, learning conditional distributions to generate plausible values
-
Machine Learning: Developed predictive models, including Gaussian Process regression, to estimate material properties from industrial process parameters
-
Statistical Modeling: Explored and benchmarked statistical and ML approaches for incomplete, heterogeneous industrial datasets
-
Predictive Modeling: Investigated relationships between process conditions and material-property profiles to support data-driven estimation
LEARNINGS
This experience significantly strengthened my expertise in Exploratory Data Analysis (EDA), Statistical Modeling, Machine Learning, and Generative AI, particularly when working with complex, incomplete, real-world industrial data.
Beyond technical skills, working in an R&D environment developed my critical thinking, scientific rigor, and independent problem-solving abilities. I learned to question assumptions, evaluate model reliability, and balance predictive performance with physical consistency and practical industrial relevance.
For confidentiality reasons, I'm not allowed to expose the results and the developpement of my work
TRATO
CONTEXT
This internship was carried out at Trato-TLC, a French company specialized in the design and manufacturing of lighting systems and luminaires. I joined the company in January 2025 as part of the mandatory stage ouvrier required by École Centrale de Lille. The objective of this internship was to gain firsthand experience with manual and operational work, in order to better understand the realities of production environments and the people involved in them.
WHAT I DID
My role was focused on hands-on production work related to the assembly of lamps and lighting fixtures. This involved practical tasks such as assembling electrical circuits, handling electronic components, wiring, and testing finished products. I regularly worked with tools and equipment such as screwdrivers, drills, measuring instruments, and various fastening systems, which required precision, attention to detail, and respect for safety procedures. Through this experience, I developed a concrete understanding of how theoretical concepts in electronics translate into real, physical products.
LEARNINGS
Beyond the obvious gain in practical knowledge of electrical circuits and electronic components, this internship had a significant impact on my communication skills in French. Working in a fully French-speaking environment required constant interaction, asking questions, clarifying instructions, and engaging socially with colleagues. As a result, my French improved dramatically over the course of the month, both technically and conversationally.
Equally important, this experience gave me a deep appreciation for manual work. Being directly involved in production highlighted the skill, precision, and responsibility required to transform designs into reliable physical products. It reinforced my respect for the craftsmanship behind engineering solutions and reminded me that high-level design only has value when it is grounded in well-executed hands-on work.
bottom of page