SoftBytes
Plan Your Project
AI Automation & Agents

Private Q&A over investment notebooks

A natural-language Q&A interface, built on Claude through Amazon Bedrock, over proprietary investment notebooks and financial content stored in S3.

  • Sector Financial services, Wealth management
  • Client Turnkey asset management platform
Claude served through Amazon Bedrock
Bedrock
Answers drawn from notebooks and documents in S3
S3-backed
Django APIs, with a Streamlit interface on top
API-first
The starting point

The challenge

The team needed answers from proprietary investment notebooks, market material and investment documentation, but finding them meant searching a large document repository by hand.

The solution

What we built

  • Built a Q&A service on Claude through Amazon Bedrock that answers questions from the platform's own notebooks and documents.

  • Connected the S3-backed workflows where the notebooks and related financial content are stored.

  • Exposed the service through Django APIs, with a Streamlit interface for asking questions.

Architecture

How the system works

How it fits together
  1. 01QuestionPlain-language query
  2. 02Django APIRequest handling
  3. 03DocumentsNotebooks and content in S3
  4. 04Claude on BedrockAnswer generation
  5. 05AnswerIn the Streamlit interface
Scope

What was delivered

01

Q&A service on Amazon Bedrock

02

S3 document workflows

03

Django APIs

04

Streamlit interface

Technology

Stack

Amazon BedrockClaudeAmazon S3DjangoStreamlit
More work

Related case studies

See all AI & Automation case studies

Need something similar?

We can scope and build a system with comparable capabilities, shaped around your own workflows, data and integrations.

Plan a similar project More case studies