Documentation Index
Fetch the complete documentation index at: https://docs.bigdata.com/llms.txt
Use this file to discover all available pages before exploring further.
May 22, 2026
Widgets, Chat Widget, Document Viewer/5 minutes read
Introducing Bigdata Widgets: a one-line embed that puts our Research Agent inside any app, on any stack, with your brand on it.
Every team building with AI eventually hits the same wall. You have the model. You have the data. You have the API key. And then you spend the next six weeks wiring up streaming, building a citation UI, taming markdown rendering, debugging token-by-token updates, and arguing about whether the chat bubbles should be rounded on all four corners or just three. The interesting part of the work (the research, the insights, the answers your users actually came for) gets buried under frontend plumbing. We built the Chat Widget and the Document Viewer Widget to remove that wall entirely.
| Chat Widget | Document Viewer Widget |
|---|---|
| Embed a fully functional AI research assistant into any web application with a single code snippet. Handles streaming, attribution, and branding out of the box. | Embed rich document experiences that load sources cited by the Research Agent, scroll to the exact referenced paragraph, and highlight the relevant content automatically. |
The Chat Widget
The Chat Widget makes Bigdata’s Research Agent embeddable in any web application with a single code snippet. Drop a<div>, drop a <script>, and you have a fully functional, fully branded AI research assistant running in production.
No streaming logic to maintain. No attribution layer to build. No design system to reconcile. Just an embeddable surface that handles all of it for you.
What you get out of the box
| Zero-build deployment | Auto-generated embed code ready for HTML, React, Node.js Proxy, and Python Proxy: copy straight from the platform, adjust, and ship. |
| Full white-label branding | Swap in your logo. Tune every color token: Primary, Background, Surface, Text, Border, with hex inputs. Set your font family and border radius. Toggle the “Powered by Bigdata” badge on or off. |
| Inline source attribution | Every answer traces back to its source: earnings calls, filings, news, or whatever you’ve scoped it to. Citations render inline so users see exactly where insights come from. |
| Guided starter prompts | Pre-populate clickable starter prompts so users immediately understand what to ask. New users hit the ground running instead of staring at an empty input box. |
| Live preview | Tune everything visually on platform.bigdata.com and verify every change before shipping. What you see in the playground is exactly what your users will see. |
The Document Viewer Widget
The Document Viewer Widget lets you embed rich Bigdata document experiences directly inside your own sites and applications: no redirects, no context switching, no loss of workflow. It is the natural complement to the Chat Widget. When the Research Agent answers a question and cites a source, the Document Viewer Widget completes that workflow. Users can click any citation, load the underlying document, and read it in-context, all without leaving your application. But the viewer goes further than simply loading a document. It automatically scrolls to the exact paragraph referenced by the Research Agent and highlights the specific content that was used to generate the insight. Users don’t have to hunt for the relevant passage: they land on it instantly.Key Capabilities
| Automatic scroll-to-reference | Opens the document at the precise paragraph cited by the Research Agent, with no manual searching required. |
| Content highlighting | The exact text used by the Research Agent is visually highlighted within the document, making the evidence immediately verifiable. |
| In-app document reading | Documents load and render inside your application. Users stay in your product rather than being sent to an external viewer. |
| Independent or connected | The Document Viewer can be deployed on its own for standalone document experiences, or connected to the Chat Widget to create a seamless end-to-end research workflow. |
| Same embed simplicity | Like the Chat Widget, the Document Viewer is a single code snippet. No backend required; no custom rendering to maintain. |
Together: A complete research workflow
The two widgets are designed to work independently, but connecting them makes the difference between a good AI experience and a great one.A user asks a question in the Chat Widget. The Research Agent responds with an analysis and inline source citations. The user clicks a citation. The Document Viewer Widget opens the source document, scrolls directly to the referenced paragraph, and highlights the exact content the Agent used.The full loop (question, answer, evidence) happens inside your application. Users never lose context. The experience is coherent, credible, and entirely within your brand.
Who this is for
Research and investment teams who want a branded research assistant in their investor portal or client dashboard without commissioning a six-month engineering project. SaaS platforms adding AI-powered document research as a feature, without staffing a team to maintain streaming and attribution logic forever. Internal analyst tools that need earnings, filings, and news intelligence scoped to a specific universe of sources and workflows. Sales and product teams building demos for prospects or stakeholders before committing real engineering resources. Configure it in an afternoon, demo it the next morning.Why it matters
There is a real cost, measured in months, not weeks, between having access to a powerful research model and shipping it to end users. For most teams, that cost is the reason “AI features” stay stuck in the roadmap section labeled next quarter. The Chat Widget collapses that gap. The frontend work that used to stand between an API key and a production-ready AI research experience is now handled. Configure visually. Preview live. Deploy in minutes. The engineering team gets to focus on the parts of the product only they can build. The research team gets a fully functional assistant their users can actually use today.Try it
If you’re building anything that involves users asking questions of financial documents, news, or filings, and you’ve been quietly dreading the frontend work, this is for you. What used to be a sprint is now a copy-paste. Reach out at sales@ravenpack.com or visit www.bigdata.com to see the Chat Widget in action. Full documentation for both widgets is available in Bigdata docs. If you already have access to Bigdata, head to platform.bigdata.com/widgets. The playgrounds there mirror what you see in the screenshots above: configure both widgets, preview them live, and grab living embed snippets you can plug and play into any stack, with no separate build step required.
Víctor Pimentel Naranjo
Senior Product Manager, Team Lead
Raul Balanzino
Product Manager (VP)