In the silent ocean of the Galaxy, mature civilizations have likely long chosen an unexpected strategy: instead of loud beacons, they scatter quiet fireflies — autonomous interstellar laser beacons, drifting in the void and dropping microsecond flashes. These 'seeds of contact' are not tied to their home stars and do not drown in their glow. For six decades, SETI relied on radio waves, the echo of Karl Jansky's discoveries, but technological progress invites us to change the orchestra: shift from listening to cosmic noise to observing its rhythmic twinkling.

The calculation is pragmatic: why build a transmitter that devours a star's energy when you can seed the interstellar medium with a swarm of cheap probe-beacons? Each is a laser with the power of Earth’s energy system (~10⁸ W), capable of living for millennia. A microsecond flash, squeezed by diffraction into a narrow beam, looks from 20 parsecs like a spark drowning in the photon ripples of the night sky. But unlike radio waves, smeared by interstellar plasma, an optical pulse arrives almost undistorted — an instant imprint of alien technology, crossing the abyss of centuries. When there are millions of such flashes and they repeat 413 times a second, their pattern can be fished out of the noise.

Modern optics and computing make such a search real for the first time. The 'Space Beacons' project combines three breakthroughs. First, ultra-light diffractive-refractive lenses (MODE) with a wide field of view. Second, cameras on fast CMOS detectors that count single photons. Third, machine learning on GPUs, processing terabytes of data in real time. A meter-class telescope, scanning the sky with 10-microsecond resolution, looks not for the flashes themselves, but for their pulsation. A folding algorithm stacks exposures synchronized with an assumed period: this 'folding' of a light curve accumulates signal proportionally to the square root of the number of pulses. By accumulating ten million flashes (just one hundred seconds of pure glow), we achieve a signal-to-noise ratio sufficient for confident detection.

Calculations based on numerical models and analytics paint a tantalizing forecast. If a beacon intentionally targets our system, a ground-based telescope with a meter mirror can detect its twinkling from distances up to 20–22 parsecs. In this sphere swarm about three thousand stars, and many are surrounded by exoplanets. A negative result will also be a discovery: for the first time, it will impose a strict limit on the number of technological neighbors in near space, adding fuel to the debate on the Great Silence. A positive one would mean we've caught not a random shout, but the steady beat of a cosmic heart.

In the coming years, the initiative could grow from student projects into a global network of robotic telescopes. Along the way, a new astronomy will be born — the science of the millisecond optical Universe, where not only artificial beacons hide, but also fast transients of unknown nature. The ideas of Charles Townes on laser interstellar communication get a second wind: perhaps space has long sparkled with rhythmic light, and all we need to do is tune the receiver.