$ data labeling for ML— training annotations with quality control and double-checking
$ knowledge graph— entities and relations from your data, a base for RAG and search
$ anonymization & pseudonymization— anonymization takes data out of GDPR scope; pseudonymization reduces risk but data stays in scope — we match the method to the use case
$ data quality monitoring— rules and alerts for anomalies before they reach a report or model
$ custom scraping— many sources, within ToS and the law (we check the legal basis first)