Engineering at Nutrient (@nutrient_eng) header
Engineering at Nutrient (@nutrient_eng) profile picture

Engineering at Nutrient is where we share insights into how we’re building the @nutrientdocs SDKs, Document Web Services APIs, and AI capabilities in documents.

Raleigh, NC USA
Joined June 2025
Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)Photo by Engineering at Nutrient (@nutrient_eng)
AI answers are useful. But answers your software can actually act on exponentially more powerful. Meet DocSignals: Nutrient document extraction + Jev, TypeSafe’s AI model. Ask a question about a document. Get a structured signal and a citation you can inspect. Then define what happens next: route the document, trigger an automation, or bring in a person to review. Probabilistic signals. Deterministic workflows. Explore the open-source project and test docsignals for free -- no API keys needed: bit.ly/4rkBv1P
1
35
OCR makes documents readable. It doesn't make them usable. Nutrient Data Extraction API parses PDFs, scans, and forms into structured, source-grounded data you can actually use. Nutrient Data Extraction API is built for AI, automation, and compliance teams.
1
17
There’s a lot that goes into generating a PDF that you don’t think about until you have to build it. How will the layout handle different amounts of data? Who will maintain the templates? What happens when you’re generating thousands of documents? Those details matter when choosing an API. We compared six PDF generators, looking at how they handle templates, pricing, and the work that happens after a PDF is created. Read the comparison: bit.ly/48iGqIy
1
23
Ever wondered how software reads a scanned invoice? There’s a lot of work hiding between “we received the invoice” and “the data is ready to use.” OCR invoice processing helps automate that work. Optical character recognition reads text from scanned documents, so details like invoice numbers, dates, and amounts can be extracted and used. This blog explains how the process works and what affects accuracy. Read the blog: bit.ly/4rb8PZb
11
AI can finally read bad handwriting. Just one problem: most models won't tell you where on the page they read it, or how it relates to the other elements on the page. Nutrient's Vision API adds word-level bounding boxes and confidence scores that finally give your agents all the context they need. Grounding matters more than accuracy. Learn more about Nutrient's handwriting recognition philosophy: twp.ai/4huNDa
10
In 2026, people don’t want to use your product. They want their agent to use it for them. Developers describe the outcome. Their agent reads the docs, finds the APIs, and writes the integration. It's really that simple. That’s why we’re building Nutrient as an agent-first document platform. We open sourced agentic-usability to measure how well Claude Code, Codex, and Gemini can use any SDK, including ours. Your next customer may never read your documentation. Their agent will.
1
15