Case studies/Omno AI
Computer vision as a product, not a lab demo
Shared vision services for retail, sports and traffic — one pipeline, different jobs, a console a customer can log into.
- Client
- Omno AI
- Location
- United States
- Year
- 2025
AI · SaaS
info@lyrocode.com
Shape
Vision SaaS
Verticals
3
Year
2025
Pictures from the work
Photography that matches the product — the buildings, the floor, the clothes, the road — not stock of people pointing at a laptop.
The job
The models worked in a notebook. Customers needed a product: upload or stream, a job, a result, a bill. Retail, sports and traffic share detectors but not the report the customer wants to see.
Each vertical was at risk of becoming a fork. We had to keep one pipeline without making every dashboard a compromise.
What we built
A common inference and job layer, then thin vertical apps on top — counts for a store, events for a venue, vehicles for a corridor. The SaaS console is where customers create sites, cameras and exports.
We treated evaluation as part of the product: a clip that fails QA does not silently become a number. Same people on the model and the API, so the console does not lie about what the network can do.
Where it landed
Omno sells a console, not a research drop. New camera jobs reuse the pipeline. Vertical reports stay specific without three codebases to keep alive.
- Shared inference and job pipeline
- Vertical reports for retail, sports, traffic
- Customer console for sites and cameras
- QA on clips before numbers ship
Stack
What we shipped with — picked for this product, not for a slide.
Python
PyTorch
TensorFlow
FastAPI
React
AWS
Docker
PostgreSQL
“We had models. We did not have a product. They sat with us until a customer could log in, run a camera, and download a report without calling an engineer.”
More work
Willis Bond·Web · Property
Project sites a development team can actually keep current
Marketing and project pages for a New Zealand developer — precinct photography, availability, and content the marketing team can change without a ticket.
Read the story
Maria B.·Commerce · Web
A fashion storefront that has to survive sale season
Commerce work for Maria B. — lookbooks, collections, and a checkout path that still holds when a drop hits.
Read the storyTell us what you are building.
We usually reply within one business day. Email info@lyrocode.com if you would rather skip the form.
