SaFuture Inc & Qwhery

Work, 2026

Background

SaFuture Inc and Qwhery are two startups focused on developing IT and GIS solutions for counties and municipalities.

What I Did

The internship was split in three parts with the intent goal of
1. creating an improved data indexing pipeline
2. automating sales prospecting
3. on knocking out a long list of feature requests for a client.

Chatbots for Property Assessment

I worked on improving the existing data indexing pipeline by integrating batch processing and adding better integration into the web dashboard.

The backend involved over 160 API routes to manage all of the moving parts involved with the product. I contributed to various parts of this, as well as writing new routes for control over data indexing. This was a unique challenge as changes would require simultaneous modifications across three microservices.

Here is a video of the chatbot in action, deployed for Rutherford County, TN:

Sales Lead Identifier

The existing flow for identifying sales prospects was to manually web search government directories for leads. Thus, automation had huge potential benefit here.

Automation had been previously tested using web scraping, but yielded low accuracy results. I experimented with web search APIs, Playwright web scraping, and LLMs with web search like gpt-4o-mini-search-preview. LLMs proved to be the most efficient. I integrated the Hunter.io API for email search and LangChain for orchestration.

The platform achieved 80% accuracy and saved 30+ hours of manual searching per week. This was my overall favourite project as I got to build it from scratch and pilot it.

Work for Clients

I can't disclose as much for this one, but I shipped 25+ feature requests. This project was interesting, and I learned how to manage stakeholders who are in charge of different parts of the tech stack.

• UI improvements and bug fixes across map dashboards
• New API endpoints and database queries for client-specific data and features such as notes
• Improving usage of Google Maps API for better map indicators

The Experience

I integrated into the team's workflow pretty quickly, shipping code from day two. We were given a lot of ownership over our projects and I learned a lot about both software engineering and mapping technologies.

Takeaways