Projects · Engineering case studies

Projects

Detailed engineering breakdowns of my embedded and firmware work: architecture, technical challenges, implementation, and the engineering decisions behind each build.

01Hardware

In Progress

Custom Sensor Telemetry Board

Custom 2-layer STM32F411 PCB with FreeRTOS environmental telemetry

A custom-designed 2-layer PCB built around an STM32F411 that samples a BME280 environmental sensor over I²C and streams telemetry over USB-C, with FreeRTOS firmware coordinating acquisition and transmission.

MCU: STM32F411 (LQFP-48)
Board: 2-layer · 50 × 40 mm
Sensor: BME280 @ I²C 400 kHz
Link: UART 921600 baud · USB-C

STM32F411FreeRTOSKiCadBME280I²CUART

View case study

02Firmware

In Progress

Multi-Peripheral SPI Telemetry System

Bare-metal shared-bus SPI driver with a FreeRTOS telemetry pipeline on STM32F4

A bare-metal SPI driver on the STM32F446RE that services multiple peripherals on a shared bus with independent chip-select control, feeding an interrupt-driven FreeRTOS producer/consumer telemetry pipeline with fault recovery and dropped-frame detection.

MCU: STM32F446RE
Bus: SPI1 @ 1 MHz · shared, multi-CS
Model: Producer / consumer + ISR
Safety: Timeout recovery · seq. validation

STM32F446REBare-Metal CFreeRTOSSPINVICISR

View case study

03Firmware

Completed

STM32 Bare-Metal Peripheral Driver Library

A bare-metal embedded firmware library for the STM32F4 that implements UART, I²C, and SPI by directly configuring RCC, GPIO, and peripheral registers. It includes an interrupt-driven UART receive path and a validated BMP280 sensor integration over SPI.

Core: ARM Cortex-M4 (STM32F4)
UART: 115200 baud · IRQ RX
I²C: 400 kHz Fast Mode
Sensor: BMP280 ±0.5 °C / ±1 hPa

STM32F4ARM Cortex-M4CUARTSPII²C

View case study

04Firmware

Planned

Secure STM32 Bootloader

Planned: robust in-application firmware update with integrity checks

An upcoming bootloader project for STM32 focused on safe, recoverable firmware updates: in-application command interface, controlled bootloader entry, CRC-validated images, corrupted-image detection, and boot-loop prevention.

Status: Planned
Entry: Controlled, on next reset
Integrity: End-of-flash CRC
Safety: Boot-loop prevention

STM32BootloaderCRCFirmware UpdateReliability

View case study

05Applied

Completed

Multi-Modal Patient Distress Detection System

Audio-triggered patient monitoring on Raspberry Pi 5 pairing FFT voice analysis with DeepFace emotion recognition

A real-time patient-monitoring system built on the Raspberry Pi 5 as part of a 5-person team. An audio stage runs FFT spectral analysis of the patient's voice and, when triggered, escalates to DeepFace facial-emotion classification through a multi-frame voting state machine, reaching roughly 1 second trigger-to-alert latency and 90% detection accuracy.

Platform: Raspberry Pi 5
Audio: FFT · 85–2000 Hz bands
Vision: DeepFace emotion
Result: 90% acc · ~1 s alert

Raspberry Pi 5PythonDeepFaceSciPy FFTOpenCVComputer Vision

View case study

Note: projects marked “In Progress” are under active development; case-study pages document the design and current implementation. Media (PCB renders, scope captures, diagrams) is being added as it is produced.

Open to opportunities

Seeking Fall 2026 embedded / firmware / hardware co-op.

If you're hiring for embedded, firmware, or hardware roles, I'd be glad to talk through my projects and how I work. The fastest way to reach me is email.

m46saad@uwaterloo.ca Contact page

Email LinkedIn