Deeds, court filings and patents
Public records with centuries-old conventions — legal descriptions, docket structure and priority dates.
Public records are the oldest structured data most organisations touch, and they are structured in a way no schema designer would choose. A deed’s operative sentence is a survey call written in compass bearings. A case caption is a two-column table drawn with a column of closing parentheses. A patent front page is keyed by two-digit numbers in brackets, because the label beside them might be in German. None of this is legacy formatting that a good model can normalise away. It is the record, and in a dispute the record is what the document says, not what your pipeline made of it.
These pages are about the specific thing that goes wrong in each family of document — the stamp that lands on top of a name, the deadline that is a rule rather than a date, the claim whose meaning is carried by the number of the claim it refers to. The general machinery of extraction is covered elsewhere in this library; what is here is the part that only applies once you know which document you are holding.
Extracting the Legal Description From a Property Deed
Why a legal description is captured as one verbatim block first and parsed into fields second, and what breaks when that order is reversed.
10 min read
Extracting Grantor and Grantee Names From a Deed
How to get the direction of a conveyance right, and what to do when the county's recording stamp lands on top of a party name.
10 min read
Extracting Recording Information From a Scanned Deed
Why a book-and-page or instrument-number reference is only an identifier together with its county and book series, and how to normalise without destroying the citation.
10 min read
Extracting Case Numbers and Docket Entries From a Court Filing
Why the caption block is the worst place to read a case number from, and what the electronic filing header gives you instead.
10 min read
Extracting Motion Type and Filing Party From a Court Document
How to get the document's own type from its title rather than from body text that names the other side's motion, and where the filing party is actually stated.
10 min read
Extracting Hearing Dates and Deadlines From a Court Filing
Why a relative deadline should be stored as a trigger, a period and a rule set rather than as a date, and what the computation rules actually say.
11 min read
Extracting Claims and Priority Dates From a Patent Document
Why claims form a dependency tree that has to be preserved by number, and why a single scalar priority date is the wrong shape for the answer.
11 min read
Extracting Inventor and Assignee Information From a Patent Filing
How INID codes make a patent front page readable regardless of the language it was printed in, and why the printed assignee is not the current owner.
10 min read
Extracting Cited Prior Art From a Patent Document
Why the references-cited block is three differently shaped lists printed as one, and how to type each citation instead of flattening it to a string.
11 min read
Other topics
- LLM fundamentals & architecture
- Tokens, tokenization & context windows
- Prompt engineering
- Reasoning models & test-time compute
- Multimodal AI: vision, audio, video
- RAG & retrieval
- Embeddings & vector search
- AI agents & tool use
- Structured output & function calling
- Fine-tuning & post-training
- Local inference errors, string by string
- Running local models day to day
- Testing code that calls an LLM
- Snapshot and property testing for model output
- Regression suites for prompts
- Eval gates in CI
- Flaky tests against a model
- Determinism and the cost of testing
- Contract and streaming tests
- Testing tool calls and retrieval
- Inference, serving & latency
- Rolling out a prompt change
- Testing AI systems in practice
- Forecasting a time series
- Machine learning on tabular data
- Geospatial data and models
- Understanding audio that is not speech
- Understanding video
- Core computer vision tasks
- Machine learning on graphs
- Point clouds and 3D
- Evaluation, benchmarks & LLM-as-judge
- Sensor and IoT data
- Logs and event streams
- Models over biological sequences
- Machine learning on molecules
- Embedding and searching code
- Extracting invoices and purchase orders
- Receipts, statements and tax forms
- Insurance policies and contracts
- Extracting from medical records
- Observability & LLMOps
- CVs, certificates and identity documents
- Shipping, customs and technical documents
- Meetings, email, chat and filled-in forms
- Building an extraction pipeline
- Business, property and inspection documents
- Contract clauses and insurance claims
- Regulated and compliance documents
- Consumer, travel and closing documents
- Mapping one chat API onto another
- SDK and framework migrations
- Hallucination & failure modes
- Re-embedding and model deprecation
- Cutting over between providers
- Parity gaps, shims and legacy endpoints
- Moving between model versions
- Migrating vector stores and caches
- Mapping capabilities and parameters
- Migrating pipelines and agents
- Contracts, runbooks and rollback
- Auditing a codebase before a cutover
- Compliance and fine-tune migration
- LLM cost engineering
- Routing, cost tracking and multi-tenancy
- What a migration does to your prompts
- AI security & prompt injection
- Privacy, compliance & data residency
- AI governance, policy & society
- Building reliable AI applications
- AI hardware, GPUs & compute
- Open-weight models & local inference
- AI for developers & coding agents
- AI in industry: vertical playbooks
- AGI, superintelligence, alignment & the long future
- Machine learning foundations
- NLP fundamentals & classical tasks
- Data engineering for AI
- Synthetic data & dataset curation
- AI product design & UX
- Search, ranking & recommendation
- Enterprise adoption & change management
- AI careers, skills & teams
- Reading AI research
- AI in science & discovery
- Robotics & embodied AI
- AI economics, markets & business models
- AI myths, hype & media literacy
- Context engineering
- Shipping AI features: patterns & anti-patterns
- Build it: end-to-end AI tutorials
- Python for AI: hands-on recipes
- TypeScript, React and the web
- Frameworks and SDKs
- Errors and troubleshooting
- AI facts, numbers and statistics
- The history of AI
- The maths behind AI
- Architectures beyond the transformer
- Reinforcement learning
- Diffusion and generative media
- Speech, audio and voice engineering
- Benchmarks, one at a time
- AI search visibility
- Infrastructure and operations
- Databases and storage for AI
- Knowledge graphs and structured knowledge
- Classical ML in production
- Regulation, jurisdiction by jurisdiction
- Prompt recipes and pattern library
- AI for people who do not write code
- Writing, media and creative work
- Edge and on-device AI
- Interpretability and model internals
- Field notes
- OpenAI model behaviour
- Claude model behaviour
- Gemini model behaviour
- Llama model behaviour
- Mistral model behaviour
- Qwen model behaviour
- DeepSeek model behaviour
- Cohere model behaviour
- Grok model behaviour
- Small model behaviour
- Hybrid model architectures
- Token cost by language and script
- Transliteration, romanization and script handling
- Locale-correct output
- Multilingual generation quality
- Multilingual pipelines
- The EU AI Act, article by article
- AI under the GDPR and EU data law
- US AI regulation, state and sector
- International AI governance and standards
- AI litigation and enforcement
- Running AI workloads on AWS
- Running AI workloads on Google Cloud
- Running AI workloads on Azure
- AI at the edge: Workers, Vercel and Netlify
- Serving models on Kubernetes
- Operating AI infrastructure
- Quantization formats and what they cost
- llama.cpp, flag by flag
- Ollama and the desktop local-model runtimes
- Local models on Apple Silicon
- Hardware for local inference
- Running speech and embedding models locally
- Model files, adapters and conversion
- VRAM arithmetic for local models