For 20+ years, we've all designed APIs for developers. At Unstructured, we've spent the last year designing them for a different consumer entirely — and it's changed how we think about software. The caller isn't a person reading docs anymore. It's a model, reading your tool descriptions as part of its prompt, in the middle of a task, with a finite context window and no patience for ambiguity. Building our Transform MCP server — the skeleton key utility that lets agents like Claude, ChatGPT, Grok, etc. turn any of 60+ document formats into universal structured data they can use for building reliable business automation, RAG for powering virtual assistants, or semantic memory for agents— taught us so many lessons about the emerging new agentic era status quo. Your documentation is now your interface. A tool description isn't reference material a human skims once. It's read on every single call. Every sentence either steers the model correctly or costs a failed job. We render our descriptions from the same source of truth as the code, because drift a human would shrug off will send a model confidently down the wrong path. Error messages are a part of the conversation. When a model hits a limit, the error text is the only remediation path it has. Write it for the reader who can actually act on it. Every token is UX. We learned the hard way that a "complete" response full of base64 payloads isn't generous — it's context pollution. Restraint is the hardest discipline. The temptation is to add a tool for every capability. But every tool is cognitive load in the prompt. The craft is folding new power into vocabulary the model already knows. The deeper realization: most of the world's knowledge still lives in documents — PDFs, decks, contracts, emails. The agentic era won't reach its peak until that dark matter becomes structured, and the interfaces that do it are being designed right now. We're not writing APIs anymore. We're writing instructions for artificial minds.

The Document AI space has seen a fundamental shift in the past year. Everyone—from scrappy startups to established players—has pivoted from custom supervised models to wrapping the same handful of closed-source multimodal models. Yet, despite the fact we're all using essentially the same approach and the same models under the hood, there's no shortage of “benchmark triumphs” from Document AI vendors touting the best performance on the market. I especially find it comical when these vendors compare their product against ours at @UnstructuredIO , and yet instead of comparing their VLM wrapper against our VLM wrapper (which according to our own benchmarks outperforms theirs), they compare it to our free, open source product—a product that doesn't depend on massive, powerful, expensive closed source models. *blink blink* I'm sorry, but that's like comparing public transportation in Rome to driving an Alfa Romeo 4C Spider convertible through the Tuscan hills—they were designed for different intents in mind. Here’s the truth: when Fortune 500 teams run real head-to-head evaluations—our commercial platform consistently performs on par or better than the best in the business. Month to month, we trade #1 spots with the leaders. But the bigger problem is this: benchmark theater is costing enterprises greatly. Choosing a vendor that is touted via their own benchmarks as having the best-in-class transform of pdfs, but can't process other document types results in organizations having to build a rats nest of supplemental home-grown capabilities that require management, maintenance, and eventually grows to the point where it needs to be swapped out with a more scalable solution. Those glossy accuracy charts usually measure PDFs in isolation—while critical data in .docx, .pptx, .eml, .msg, .tiff, .epub, or .xlsx files goes completely unseen. And what about model fallback, dynamic content-based routing, retries, and all the other features needed to ensure your VLM wrapper actually works at scale? Finally, let's not forget the factor that when it comes to benchmark performance, most vendors fine-tune (to the point of overfitting) prompts to perform well on major public benchmarks. At the end of the day, document transformation quality isn’t about cherry-picked metrics. It’s about coverage, fidelity, metadata richness, and mitigating the cost of missed information. Ready to see what benchmarks look like when they reflect real business impact? 🎙️ Join our deep-dive in next week's webinar on Wednesday, Sept. 10: Document Transformation Quality Series: Pushing the Boundaries of Document Transformation Quality → Sign up here: unstructured.io/events/pushi… #DocumentTransformation #BenchmarkTruth #EnterpriseAI #UnstructuredData #DocumentAI #Unstructured #BenchmarkData

At @UnstructuredIO, we often get the question "how well do you perform on scanned forms that include handwriting?" These types of documents are notoriously among the most difficult types of documents to ingest cleanly and reliably, yet they remain ubiquitous across many industries and are especially prevalent in healthcare, insurance, and similar domains. Our short answer? Brilliantly. But we encourage you to see for yourself via our free trial! → unstructured.io/?modal=try-f… Our industry-leading VLM partitioner is designed to tackle the most complex documents generally across all business domains, but it is especially powerful when it comes to scanned, rotated/skewed, and/or handwritten documents. Parsing these documents with less sophisticated parsers results in one or more of the following: strings of jibberish characters due to inaccurate OCR; signatures treated as blobs; form fields lost; checkboxes ignored; marginal notes dropped entirely; or worse. By leveraging state-of-the-art models and grounding our VLM partitioner in a rich document element ontology, we produce rich, clean parses of these documents, without collapsing the document's structural context: - Handwritten fields captured as structured inputs with handwriting transcribed - Checkboxes encoded as checkboxes, not flattened text - Signatures and logos preserved distinctly - Page numbers and layout context retained - Layouts and sections captured The result: even your most complex, analog-origin documents are parsed into a consistent, auditable structure that downstream systems (data entry, RAG, compliance, analytics) can trust. See an example below: a scanned, tilted, complex, medical form, filled in by hand with dummy data on the left and our parsed, rendered, stylized HTML on the right. Of course, when VLMs and handwriting are concerned, very few parses will be 100% perfect, but even for complex, messy forms like this, you can often expect very high 90s in terms of both layout and textual content accuracy from our partitioner. This example evaluated at ~98+% for both content and layout accuracy. Want to learn more? Join us for my upcoming webinar: Document Transformation Quality Series: Pushing the Boundaries of Document Transformation Quality - unstructured.io/events/pushi… #DocumentAI #Handwriting #ScannedDocs #VLM #Ontology #DataQuality #ScannedForms