Firmware
A fleet of connected devices is a relentless text generator. Every device runs a dozen cooperating services, and each one narrates its life in log lines: startup banners, retries, sensor readings, stack traces. Across a hundred-plus devices that adds up to hundreds of thousands of lines a day, spiking into
Hardware
Timing discipline during voltage transitions Power does not become stable in one instant. A board may receive input power, but the internal rails are still ramping. A regulator may be enabled, but its output has not yet reached tolerance. A level shifter may see one side alive before the other.
Most production incidents do not begin with faulty application code. They begin with infrastructure changes that nobody fully understood. A security policy modified through a cloud console. A manual configuration update made during an urgent fix. An environment that slowly diverges from the documentation intended to describe it. Individually, these
Product
In product development, performance improvements are often associated with new electronics, better sensors, or more sophisticated algorithms. Occasionally, however, the solution comes from something much simpler. A few millimeters of geometry. During the development of one of our wearable tracking devices, we encountered an issue that initially appeared to be
Firmware
A thermal camera can see a fever before a dog acts sick — but only if you know which warm pixels belong to the dog. A low-resolution thermal sensor and a normal camera look at the same scene from slightly different places, through different lenses, at different resolutions. Point them at
Firmware
A barking dog tells you something. A lunging, barking dog tells you a lot more. On a battery-powered pet collar, the two most informative senses are motion and sound — but you can't run a microphone flat-out all day, and a motion sensor alone can't tell a
Hardware
Designing perceived responsiveness while respecting electrical reality Users rarely judge a product by how fast the hardware actually becomes fully ready. They judge it by how quickly the product appears alive. That difference matters. A device may need several hundred milliseconds, or even a few seconds, before every rail is
Firmware
There is a particular kind of failure that haunts always-on devices, and it almost never shows up in the lab. The firmware runs perfectly for a day, a week, a month — and then, somewhere out in the field, it quietly stops. No crash log, no stack trace, just a device
AI
How we are building a dog facial recognition system from scratch using landmark alignment, metric learning, and a custom-trained embedding model The Problem EverBowl is Hoomanely's precision healthcare deivce for dogs. The bowl monitors eating and drinking habits, ocular temperature, sounds and behavioral patterns over time, turning routine
Technology, Insights and Ideas.
AI
Not every safety check belongs in the same place. Some only make sense on the user’s question, some only on the model’s answer, and some on both. Here is how we structure the guardrails on our veterinary chatbot — and why the input/output split matters. A guardrail isn’
Firmware
A fleet of connected devices is a relentless text generator. Every device runs a dozen cooperating services, and each one narrates its life in log lines: startup banners, retries, sensor readings, stack traces. Across a hundred-plus devices that adds up to hundreds of thousands of lines a day, spiking into
Hardware
Timing discipline during voltage transitions Power does not become stable in one instant. A board may receive input power, but the internal rails are still ramping. A regulator may be enabled, but its output has not yet reached tolerance. A level shifter may see one side alive before the other.
Most production incidents do not begin with faulty application code. They begin with infrastructure changes that nobody fully understood. A security policy modified through a cloud console. A manual configuration update made during an urgent fix. An environment that slowly diverges from the documentation intended to describe it. Individually, these
Product
In product development, performance improvements are often associated with new electronics, better sensors, or more sophisticated algorithms. Occasionally, however, the solution comes from something much simpler. A few millimeters of geometry. During the development of one of our wearable tracking devices, we encountered an issue that initially appeared to be
Firmware
A thermal camera can see a fever before a dog acts sick — but only if you know which warm pixels belong to the dog. A low-resolution thermal sensor and a normal camera look at the same scene from slightly different places, through different lenses, at different resolutions. Point them at
Firmware
A barking dog tells you something. A lunging, barking dog tells you a lot more. On a battery-powered pet collar, the two most informative senses are motion and sound — but you can't run a microphone flat-out all day, and a motion sensor alone can't tell a
Hardware
Designing perceived responsiveness while respecting electrical reality Users rarely judge a product by how fast the hardware actually becomes fully ready. They judge it by how quickly the product appears alive. That difference matters. A device may need several hundred milliseconds, or even a few seconds, before every rail is
Firmware
There is a particular kind of failure that haunts always-on devices, and it almost never shows up in the lab. The firmware runs perfectly for a day, a week, a month — and then, somewhere out in the field, it quietly stops. No crash log, no stack trace, just a device
AI
How we are building a dog facial recognition system from scratch using landmark alignment, metric learning, and a custom-trained embedding model The Problem EverBowl is Hoomanely's precision healthcare deivce for dogs. The bowl monitors eating and drinking habits, ocular temperature, sounds and behavioral patterns over time, turning routine