99x, a worldwide know-how providers chief, continues to advance the frontier of innovation by means of its specialised 99x Yantra Labs staff, specializing in the Web of Issues (IoT), embedded programs, and sensor-based purposes. The staff’s newest developments embrace use circumstances reminiscent of sensors that information optimum photo voltaic panel placement, extending CCTV feeds to detect hearth hazards, and several other different bespoke buyer engagements.
Commenting on the staff’s line of labor, Sachith Perera, Chief Know-how Officer – Product Engineering at 99x, said, “We’ve seen a robust progress development amongst prospects who provide options built-in with IoT sensors or embedded {hardware} gadgets. A few of these platforms require sub-millisecond response instances that can not be achieved utilizing software program alone. This requires embedding the choice logic into the {hardware} and constructing customized parts to allow these response instances.”
Dilshan Jayakody, Technical Architect, heads the Yantra Labs unit at 99x. Having over 20 years of expertise in firmware growth and low-level driver design, his automated fluid degree controller was honoured by Microchip Applied sciences Journal as its featured challenge in September 2025. In October, Yantra Labs additionally hosted a buyer go to from Norwegian Subsea, the place Tonja Joseph, an Embedded Software program Engineer, was in Sri Lanka for 2 weeks to collaborate with the embedded programs and IoT groups.
Reflecting on latest improvements, Dilshan shared, “Typical hearth detection programs rely totally on smoke or warmth sensors mounted on ceilings. Whereas broadly adopted, these programs are inherently reactive as they solely set off after smoke or warmth has reached the sensors.
“Our innovation addresses this hole by leveraging pc imaginative and prescient and machine studying to detect visible cues of fireside or smoke earlier than conventional sensors reply. Processing digicam feeds on the Jetson Orin NX on the edge permits real-time detection with out counting on cloud latency.
“We’ve additionally constructed an answer to deal with the impression of shading on photo voltaic panel installations. When there may be partial or extreme shading on one panel, it creates a bottleneck that reduces general efficiency throughout related modules, leading to a ten–30% power loss relying on native circumstances.
“Our Photo voltaic Shading Analyser solves these points by means of a data-driven, site-specific strategy that optimises layouts utilizing photo voltaic information loggers geared up with GPS modules, pyranometers, and LoRa (long-range) mesh networks to gather and analyse photo voltaic depth information, modelling shading patterns to suggest the most effective panel format.”
