AI
Problem Modern pet‑health systems increasingly rely on thermal cameras to extract physiological signals such as eye temperature. The challenge is that thermal sensors are low‑resolution and almost never share the same optical center, field of view, or distortion model as the RGB camera doing the visual perception. In
Firmware
Introduction: The Decision We Didn’t Question When we started building our edge systems, the filesystem was the easiest decision we made. It was familiar, stable, and came with decades of engineering behind it. Logs went into files. State was persisted as small blobs. Recovery was something the OS already
AI
The Uncomfortable Truth About ML Projects Every machine learning project has a dirty secret: the model is usually not the problem. When a model underperforms, the instinct is to reach for a bigger architecture, more layers, or fancier training tricks. But in our experience building machine learning systems at Hoomanely,
Hardware
Introduction: Velocity Without Fragility In embedded systems, speed and stability are often framed as opposing forces. Fast iteration is seen as something you do early, before the “real” system exists. Reliability is something you enforce later, once hardware is fixed and features are locked. In practice, this separation is artificial—
AI
The Unspoken Tension of In-Home AI When we talk about smart home devices, the conversation inevitably drifts toward a single, uncomfortable word: privacy. At Hoomanely, we build technology that lives in the spaces of a pet parent's home. Our flagship product, Everbowl, is a smart feeding system that
Firmware
Filesystems in embedded systems are often treated as implementation details—picked late, configured minimally, and rarely revisited. In reality, a filesystem becomes a long-lived contract between firmware, storage media, update mechanisms, and the data pipelines built on top of it. This article reframes embedded filesystems not as storage utilities, but
Firmware
How intelligent firmware orchestrates complex device initialization for reliable embedded systems Introduction In the world of embedded systems, the moment you press power is just the beginning of an intricate dance. While hardware designers focus on voltage rails and capacitors, firmware engineers face an equally critical challenge: ensuring every component
Hardware
Introduction: Precision Is Designed, Not Assumed In hardware systems that blend mechanical structure, electronics, and sensing, consistency is not a byproduct of careful assembly—it is an engineered outcome. As Hoomanely’s vBus-based products matured into tightly integrated systems, the tolerance for variation narrowed dramatically. Mechanical alignment, sensor positioning, connector
As the Hoomanely community grew, we ran into a familiar problem. Our global feed initially worked well. It helped users discover content, ask questions, and engage with pet parents across the platform. But as activity increased, conversations started overlapping, important posts got buried, and engagement slowly shifted from participation to
AI
Introduction Transfer learning usually feels like a shortcut that just works. Pretrain on a large dataset, fine-tune on your task, and let scale do the heavy lifting. That approach worked well for us in vision - but audio broke that assumption in a very specific, painful way. At Hoomanely, we