Cortexum
Cortexum: “Applied AI studio in Singapore building custom AI agents, domain-specific models, and websites that sell. Web design and revamps for SMEs across SG and Malaysia.”
Do we actually need to train a model?
“Often not. Prompting and retrieval solve most problems more cheaply and are far easier to change later.”
View in the company profileHow much data do we need?
“Less than most teams assume, but it has to be the right data. A few hundred well-chosen, correctly labelled examples usually beat tens of thousands of noisy ones. If the data does not exist yet we help design how it will be collected and labelled.”
View in the company profileHow long does model training take?
“Four to twelve weeks. Dataset work is usually the long pole rather than the training itself, and evaluation design happens up front so there is an agreed definition of better before any run begins.”
View in the company profileWho owns the trained model and the data?
“You do. The model weights, the datasets, the evaluation sets and the training code are yours on final payment. Training runs happen in your cloud or a tenancy you control, and your data is never used to train anything for another client.”
View in the company profileCan it run on our own infrastructure?
“Yes. We deliver quantised, deployment-ready models that can run on your servers or at the edge, which is often the point for teams in regulated industries where data cannot leave the building.”
View in the company profile