AI SaaS MVP – Next.js, FastAPI, SQL Server & LLM

I am looking for an experienced full-stack developer to build a lean MVP of a B2B SaaS product using AI and business data.
The application will allow business users to ask questions about their sales data in natural language and receive answers, insights, tables and simple charts.
Initial MVP stack:
Next.js frontend
Python / FastAPI backend
PostgreSQL for SaaS configuration and metadata
Microsoft SQL Server as the first customer data source
LLM/API integration for natural-language-to-SQL and data interpretation
The MVP should include authentication, a simple chat interface, secure read-only SQL Server connectivity, semantic mapping of sales fields, AI-generated SQL with safety validation, query execution, results interpretation, simple charts, and basic query history.
This is an early-stage bootstrapped product. We are intentionally building a small, high-quality MVP rather than a large platform.
The architecture should be designed so additional databases and features can be added later.
Clean code, documentation, security and ownership of the source code are important.
Experience building SaaS products involving LLMs, databases, analytics or text-to-SQL is highly desirable.
Please briefly describe a similar AI/SaaS project you personally developed and which parts you implemented.
Important: This is a fixed-scope MVP with a limited budget. Please do not propose a large enterprise architecture. We are looking for a pragmatic developer who can deliver a clean, extensible first version.
The source code must be maintained in a Git repository controlled by the client, with clear setup and deployment documentation.