This paper presents a micro electromagnetic generator designed for vibration energy harvesting, aimed at powering wireless sensor nodes. The generator, with a component volume of 0.1 cm³ and a practical volume of 0.15 cm³, is optimized for low ambient vibration levels based on real-world application data. It uses four magnets arranged on an etched cantilever and a wound coil within the moving magnetic field. The final device produces 46 μW in a resistive load of 4 kΩ from just 0.59 m s⁻² acceleration levels at its resonant frequency of 52 Hz. The generator delivers 30% of the power supplied from the environment to useful electrical power in the load, outperforming other demonstrated examples in terms of normalized power density and efficiency. The paper includes theoretical analysis, design, simulation, and testing of the generator, as well as a comparison with other inertial generators. The generator's performance is validated through experiments using a shaker unit and accelerometer feedback, demonstrating its potential for practical applications in wireless sensor networks.This paper presents a micro electromagnetic generator designed for vibration energy harvesting, aimed at powering wireless sensor nodes. The generator, with a component volume of 0.1 cm³ and a practical volume of 0.15 cm³, is optimized for low ambient vibration levels based on real-world application data. It uses four magnets arranged on an etched cantilever and a wound coil within the moving magnetic field. The final device produces 46 μW in a resistive load of 4 kΩ from just 0.59 m s⁻² acceleration levels at its resonant frequency of 52 Hz. The generator delivers 30% of the power supplied from the environment to useful electrical power in the load, outperforming other demonstrated examples in terms of normalized power density and efficiency. The paper includes theoretical analysis, design, simulation, and testing of the generator, as well as a comparison with other inertial generators. The generator's performance is validated through experiments using a shaker unit and accelerometer feedback, demonstrating its potential for practical applications in wireless sensor networks.