Patent anatomy & jurisdictions
claims/spec/drawings priority family, legal status WIPO/USPTO/EPO feeds
Classification & retrieval
IPC/CPC keywords assignee/inventor search family de-duplication
Data model & storage
document→claim→entity links chem structures as first-class provenance & audit

Entity extraction (patents & papers)
IUPAC, trivial names, IDs targets/indications assay outcomes
Name→structure & standardization
OPSIN/RDKit heuristics tautomer/stereo handling Markush caveats
Relevance ranking & linking
claim vs spec weighting family & citation graphs assignee normalization

Prior art & novelty scanning
substructure/similarity (SAR) series-level overlap time & geography context
Freedom-to-operate (FTO) triage*
claim scope heuristics legal status snapshots outside-counsel handoff
Landscape & portfolio views
assignee share & trends target/indication maps white-space detection

*FTO outputs here are for internal R&D triage only and are not legal advice.

Build a patent-mining pipeline (ingest → extract → link → report)
Theory + Practical
Automated alerts & watchlists
assignee/IPC/CPC rules keyword & target triggers email/Slack exports
Deliverables
landscape deck (PDF) entity table (CSV/JSON) configs/notebooks for reruns