WIOWIZ began in March 2025 as a fabless team with one goal: build an edge AI chip. Not a large accelerator for a data centre, but a small, low power die meant to sit close to where data is produced. A grain silo. A border post. A dehydration line. A two wheeler. A drone.
A mix of models on one small die
The design was built to run several classes of model at once on the same die. Small models for quick decisions. Large models where accuracy mattered. Video processing models for the streams that never stop. A camera feed, a sensor, a decision, inside a tight power and cost budget. For the places we had in mind, putting that mix on one die fit the problem better than leaning on a single model somewhere else.
Open infrastructure, a modest budget
We also wanted to answer a question for ourselves. How far can a small team take a chip on open infrastructure and a modest budget? Open PDKs and open source EDA had made a lot of that path reachable. We wanted to see how much of it we could walk.
Where the tools came from
EDA was not the plan. It grew next to the chip. Each time the flow in front of us could not do something the design needed, we stopped, understood that part of the flow, and built the piece that was missing. Simulation first, because we needed to watch the design run. Then waveform debug, coverage, clock and reset domain checks, formal. One gap at a time, always because the chip in front of us required it.
Over the following months those internal tools stopped being small. Scripts became frameworks. The frameworks gained their own tests and their own regressions. The work underneath the chip started to take as much engineering as the chip.
Where WIOWIZ stood in March 2025
In March 2025, that is where WIOWIZ stood. A fabless team building an edge AI chip, and a set of internal tools built to get that chip through a flow that could not yet do everything on its own. Not an EDA company. Not a design services company. A chip, and the tools the chip kept asking for.
The next piece is the tapeout we did not finish.
Our Approach
We're building systems that think about specifications the way engineers do.
We build our own in-house EDA with an intelligence layer across it. Our stack covers the full flow,
from spec to comprehensive sign-off, on tools we build and control.
Walk-in ones, walk-in zeros