Internship for a CentraleSupélec student: framing an engineering problem

Where company experience fits in the curriculum

The generalist engineering course at CentraleSupélec lasts three years and leads to an engineering degree at master’s level. It is taught on the Paris-Saclay, Metz and Rennes campuses and organised in eight-week sequences. This block structure shapes the internship calendar: you cannot be away at just any moment without discussing it with the school.

Several forms of professional experience coexist, depending on the route chosen:

• an internship or period in the field, whose length and terms are set out in each year’s syllabus;
• a possible gap year between the second and third years;
• a semester abroad, which can take the form of a company or laboratory internship;
• a final-year project, for example as a laboratory immersion in the research track;
• the apprenticeship route, alternating between school and company.

Before looking for an organisation, work out exactly which slot your experience will occupy in the curriculum. That slot determines the length, the level of academic follow-up and the type of deliverable.

Define the problem before choosing the tools

An engineering internship may involve modelling, experiments or the analysis of an existing system. The temptation is to start with a tool (a machine-learning library, simulation software) before knowing what you are looking for. Good framing reverses the order.

1. State the question in one sentence: what should you be able to claim or do at the end of the assignment?
2. List the constraints: computing time, access to data, standards, testing budget, safety.
3. Define a measurable success criterion, however simple: a maximum deviation, a processing time, a defect rate.
4. Choose the most modest method that can answer it.

At interview, ask what resources will really be accessible on the first day: access to databases, the test bench, software and the people who know the system. Many internships go downhill simply because one access is blocked for three weeks.

Explain how what you have learned serves the stated problem

The generalist course gives a grounding in many disciplines: applied mathematics, physics, mechanics, energy, computer science and engineering sciences. This breadth is an asset but also a trap: an application that lists every course taken is hard to read.

It is better to choose two or three directly useful strengths and link them to the topic:

• a thematic-sequence project carried out with an industrial partner, where you had to set out the problem before solving it;
• a piece of modelling or programming whose assumptions you can explain;
• experience in managing a team, a student society or a start-up, if the assignment involves coordination.

For each, state what you produced and what you learned about its limits. A technical recruiter appreciates a student who can say where their method stops.

Apprenticeship, gap year or internship: comparing the options

The same career goal can take several forms in the curriculum.

The internship

It is one-off and targeted. It suits testing a sector or a profession, with a clearly identified assignment.

The apprenticeship

The apprenticeship route provides between eight and nineteen months in a company in total, alternating one month at school and one month in the company in the first year, then six months at school and six months in the company in the third year. It requires an employment contract and a search for an employer from the moment you enter the course.

The gap year

Possible between the second and third years, it allows a long experience, an entrepreneurial project or a field assignment. The school also offers an optional commitment year combining a semester in the field with six months of experience in a company, a non-governmental organisation or an institution.

No option is better than the others: it all depends on your plans. A student drawn to research, for example, will favour the laboratory project; a future entrepreneur will look at the entrepreneurship track.

Going abroad: academic exchange or internship

The eighth semester can be spent abroad in three forms: academic exchange, company internship or laboratory internship. Other schemes exist, such as full non-degree years at partner universities or double degrees, with a minimum of six months abroad for some tracks.

An internship outside France raises extra practical questions: visa or residence permit depending on the country, health and civil-liability insurance, housing, the possible taxation of your pay and recognition of the agreement by the host organisation. Begin with the mobility office’s frequently asked questions, which answer the most common queries, then confirm with the lead for your programme.

Think about the calendar too: visa and housing procedures often take longer than finding the internship itself.

Applying: where to look and how to present yourself

Companies can post internship, job or international corporate volunteering offers on the school’s careers platform, while work-study offers follow a separate section and service. It is therefore the first place to look, before the general job sites.

Complete this with:

• targeted speculative applications to technical departments, with a precise topic to propose;
• the alumni network, for example to understand how a team uses simulation or data analysis;
• the laboratories of the school and its partners, if your goal is research.

An engineering student’s CV is stronger when it presents projects with their context, their tools and a quantified result. The cover letter, kept short, explains why this problem interests you and how you plan to tackle it, without reciting the company’s brochure.

Illustrative example: a six-month internship in energy optimisation

Imagine, with rounded figures, a third-year student preparing a final-year project in an industrial company. The topic: reduce the energy consumption of a production line by 5%, a target given here purely as an example.

• Month 1: audit of existing data, meetings with operators, definition of the success indicator.
• Months 2 and 3: simplified model of the line, testing of three hypotheses on real readings.
• Months 4 and 5: on-site trials, adjustments, comparison with the starting situation.
• Month 6: report, presentation to the team and recommendations.

The quantified target may not be reached. What matters is to explain why, with measurements. The real lengths and obligations depend on the track and on the year’s syllabus.

Common mistakes

• Setting off without confirming the dates and objectives with the school’s academic follow-up.
• Accepting an assignment with no defined deliverable, which comes down to support tasks.
• Choosing the tool before the problem and ending up justifying an unsuitable method.
• Underestimating access delays to data, software or industrial sites, which are sometimes subject to security rules.
• Not planning the move abroad: visa, insurance and housing determine whether you can leave.
• Neglecting supervision: a supervisor who is rarely available weakens the assessment.

Reporting hypotheses, trials and trade-offs

Your report or presentation must make clear the starting hypotheses, the trials carried out and the trade-offs that led you to choose one solution over another, including when an avenue was abandoned. A documented dead end has value: it spares the team from repeating the same mistake.

Distinguish what comes from you from what already existed, and what has been scientifically validated from what remains an intuition. Finally, state the limits: incomplete data, particular test conditions, a duration too short to draw conclusions. This honesty is an assessment criterion as much as a guarantee of credibility in front of a future employer.