Throughout my doctoral research, I served as a teaching assistant and lab instructor across lectures (CM), tutorials (TD), and practical laboratory sessions (TP) at Sorbonne University. My teaching spans two faculties: Polytech Sorbonne (the university’s engineering school) and the Faculty of Science and Engineering. The courses range from foundational first-year physics and electronics introductions up to third-year undergraduate engineering tracks in Electrical Engineering & Automation (Licence EEA) and graduate systems courses (Master SESI).
Introduction to Electronics
First-year science students often arrive at the electronics lab having only encountered circuit diagrams on paper. In this introductory course, the goal is to make physical benchtop instrumentation second nature. Working with dual-channel digital oscilloscopes, function generators (GBF), regulated DC power supplies, and breadboards, students learn how to properly trigger a trace, isolate ground loops, and measure true RMS versus peak-to-peak voltages. The lab introduces the duality of time and frequency: using electroacoustic transducers (microphones and speakers), students observe audio waveforms in real time and watch their harmonic spectra unfold via the oscilloscope's FFT mode. Later units transition from passive networks to fundamental active components, exploring diode clipping thresholds, operational amplifier comparators, and a basic amplitude modulation (AM) radio transmission setup with envelope demodulation.
Simulation of Electronic Circuits
Before building hardware, second-year engineering students must learn how to model and interrogate circuits numerically. Using Cadence PSpice, this laboratory guides students through configuring netlists and running .OP, .DC, .AC, and .TRAN analyses to evaluate linear networks, operational amplifiers, and CMOS logic gates. Rather than relying on trial-and-error, students learn to extract precise quantities: verifying Thevenin/Norton equivalents, finding the exact -3 dB cutoff frequencies of cascaded RC filters, and using parametric sweeps to determine diode peak detector sensitivity. The exercises also confront practical non-idealities that textbook formulas gloss over, such as slew-rate and gain-bandwidth limitations in LF411 op-amps, and inductive ringing phenomena during rapid CMOS inverter switching transitions.
Numerical Methods in MATLAB
Calculus behaves very differently when forced onto finite-precision computing hardware. In this second-year methods course, students explore computational linear algebra and numerical analysis in MATLAB. The sessions begin by demonstrating how machine epsilon, catastrophic cancellation, and integer overflows can silently corrupt results. Students then code and analyze numerical tools from scratch: building polynomial interpolations via Vandermonde and Lagrange matrices to witness the Runge phenomenon, evaluating the convergence rates of composite Newton-Cotes rules (rectangle, midpoint, trapezoid, Simpson) alongside Gaussian quadrature, and formulating ordinary differential equation solvers (explicit Euler and Heun/RK2). By simulating the transient step responses of RC, RL, and resonant LC circuits, students see firsthand how numerical integration step sizes dictate numerical stability, alongside comparing direct LU and Cholesky matrix factorizations against iterative Gauss-Seidel schemes.
Scientific Programming in Python
In this programming course for second-year undergraduates, the objective is to guide students from informal script writing toward disciplined, modular algorithmic development in Jupyter Notebooks. The syllabus covers core control flow, iterative loops, and sequential collections (strings, lists, tuples, and dictionaries). Computational assignments center on practical algorithmic challenges: prime number sieving, run-length encoding (RLE) string compression and decompression, and structured text file parsing. Throughout the semester, an emphasis is placed on defensive coding practices. Every function must include a clear docstring contract, formal input assertions to prevent boundary violations, and an accompanying unit test suite to independently verify correctness before submission.
Analog Electronics
For third-year undergraduate engineers, this laboratory bridges computer simulation and physical breadboard reality. Using a closed-loop design workflow, students model semiconductor devices in LTspice—including rectifying diodes, BJTs, MOSFETs, and operational amplifiers—to establish DC operating points and frequency responses before heading to the workbench. On physical breadboards, they construct bias networks, common-emitter and common-source amplifier stages, differential pairs, and active filtering topologies. Using laboratory oscilloscopes, signal generators, and multimeters, students characterize voltage gain, -3 dB bandwidth, phase margin, and harmonic distortion. By comparing bench measurements directly against simulated curves, students learn to recognize the impact of parasitic capacitances, component tolerances, and non-ideal grounding.
Microcontroller Systems
High-level software libraries often hide the actual interaction between firmware and silicon. In this third-year lab, students work directly on the ARM Cortex-M3 (STM32F103) platform using Keil µVision5, building bare-metal embedded applications through direct register manipulation. Students establish their own hardware debug connections, configure the system clock tree, and initialize GPIO ports from the ground up without relying on prepackaged HAL libraries. Key lab milestones focus on deterministic timing: configuring the SysTick system timer for non-blocking task delays, managing nested external interrupt priorities (EXTI) with measurable latency, and synchronizing dual hardware timers to generate 40 kHz PWM bursts for ultrasonic ranging transducers while observing duty cycle accuracy and jitter directly on an oscilloscope.