Glowing Spoon

Research

The next idea starts here.

Our workbench for applied intelligence. Explore the questions we’re testing, the systems taking shape, and the lessons we’re learning along the way.

01

Research project

Dalal Ratings — verified signals from market-commentary video

An ETL and verification pipeline for extracting structured, trustworthy signals from financial market-commentary video. Given a YouTube video, the system pulls and normalizes the transcript, runs it through an LLM pipeline to extract structured calls and claims, and independently runs a computer-vision pass over sampled frames to detect and cluster on-screen signals (tickers, charts, overlays).

Explore research
02

Research project

RFP Responder — an AI-first proposal desk for enterprise bid teams

An enterprise RFP response platform built around how a real bid team actually works — not a single AI-writes-the-proposal button. A Bid Manager owns a project and assembles a team; Users draft and edit; Approvers review and sign off from a queue; a Super Admin manages the platform. Every generated artifact — a summarized requirement, a drafted response section, a generated question for the customer — moves through the same human-in-the-loop review before it counts as done.

Explore research
03

Research project

WendMe — hyper-local carpooling for verified colleagues

A trusted, hyper-local carpooling app for daily commutes — restricted to people whose work email matches a recognized company domain, so every rider and driver is a verified colleague, never the general public. A driver publishes a route and departure window; a passenger searches theirs; the matching engine scores every eligible driver on a single weighted formula — passenger walk distance, driver detour, and time-window fit — and ranks candidates accordingly.

Explore research