Industry 01
OTT & Media
2 decades of Oxagile streaming work. Your rights model, indexed once and queried.
Your rights and metadata model is too large for an agent to read, and a wrong rights answer breaches a contract. Indexed once as a graph, each question becomes a query instead of a read through the catalogue.
What changes here
Rights windows and territory logic, where an error breaches a contract
Metadata volume that defeats reading and rewards indexing
Player telemetry at a rate that makes sampling a design decision
Live events, where your rollback window is measured in minutes
AI in OTT
Where AI pays in OTT
So each use earns its place on your platform. The first 4 are work we shipped.
Work in this industry
AI work first, then the engineering underneath.
OTT & Media · Media logistics company
Highlights cut while the match is still running
A model that watches the live feed, finds the moments worth replaying, and cuts the compilation before the second half.
- 72% reduction in video processing time
- 90% accuracy identifying scoring moments and fouls
- 3.2x faster highlight compilation
- 70% increase in footage processing speed
OTT & Media · Sports technology company
Point a phone at the court, get the player
Player recognition from a phone camera during a live game, with statistics on the screen before the play is over.
- 2-3 seconds to recognise players during a game
- 43 MBps of player coordinate data generated
- 16 FPS player detection in the mobile app
- 4 person-months over 2 months to deliver
OTT & Media · AI software product provider
One AI search app across 7 TV and mobile platforms
An AI search agent for films and series, with ratings, reviews, age guidance and trailers in one place, launching each title in the service that carries it. Built as an MVP for fundraising.
- 3 months to deliver
- 7 platforms from one codebase
- 40% cost savings from cross-platform code reuse
OTT & Media · Broadcast monitoring provider
Transcoding errors forecast a month ahead
A monitoring system for more than 7,000 channels that reads early warning signs in the broadcast metrics and forecasts transcoding errors up to a month before viewers would see them.
- 7,000+ SD and HD channels monitored
- Transcoding errors forecast up to 1 month ahead
- Big data stored each month cut 10x
- Routine issue requests from end users cut 10x
Internal · Oxagile agent tooling
One architectural question, priced 2 ways
An index of how a codebase relates cut the cost of an architectural question by a factor of 25.
- ~400 tokens per architectural question, against ~10,000 before
- Under 0.3 seconds to return the answer
- 25x less context for the same question
- 700 nodes and 1,000 edges indexing one workspace
OTT & Media · Pay-TV and OTT solution provider
A pay-TV platform modernised while it stayed live
The front end of a white-label video platform, revamped feature by feature inside the vendor’s own team: new modules, a faster interface and a quieter API.
- Component re-renders cut by 50%
- API requests to the back end cut to 1 in 3
Engagements
What we take on here
The same 10 services, held to this industry's constraints.
Next step
Name the answer that went wrong.
Tell us where your rights or telemetry model last gave a wrong answer. You get a written answer on what breaks first and what it costs.