Cue: three CSE students build AI smart glasses for visually impaired people
By Thakur Prashanth · October 15, 2026 · 6 min read
G. Divyasree (BTF-010), K. Rishitha Purna (BTF-011) and Keshamoni Nikhila (BTF-012) did field research at Sri Matha Foundation for the Blind, then built Cue: AI smart glasses that detect objects, read text and currency aloud. It is being piloted with visually impaired users.
At a glance
Product: Cue
Built by: G. Divyasree (BTF-010), K. Rishitha Purna (BTF-011), Keshamoni Nikhila (BTF-012)
Background: G. Divyasree: 3rd year CSE-DS, MVSR Engineering College; K. Rishitha Purna: 3rd year CSE-DS, MVSR Engineering College; Keshamoni Nikhila: 3rd year CSE-DS, MVSR Engineering College
Cohort: Build Theory Fellowship, Cohort 01 (BT-01)
Status: Being piloted with visually impaired users.
Who built Cue?
Cue was built by three third-year CSE (Data Science) students at MVSR Engineering College: G. Divyasree (BTF-010), K. Rishitha Purna (BTF-011) and Keshamoni Nikhila (BTF-012). None of them had used GitHub or created a repository before Build Theory. Divyasree is now one of Build Theory's forward-deployed fellows.
Listening before building
The team wanted to solve a societal problem and chose to work with blind and visually impaired people. They quickly realised the community faces many different challenges and that they could not pick one without listening. With Build Theory's guidance they visited Sri Matha Foundation for the Blind and spent time with people there to understand their needs. As one of them put it: "I thought blind people would lack confidence. They all were awesome. They were more confident than us."
What Cue does
Cue is a pair of smart glasses connected to a phone. It is designed to detect objects ahead in the direction of travel, recognise text and currency, and read information aloud: an AI assistant that guides the user through audio.
The hardest part
Hardware and software integration, and training their own AI/ML models, made Cue one of the most technically ambitious projects of the cohort. The turning point came when their object-recognition model reached its target accuracy.
Where it is now
The team is piloting Cue with visually impaired people and hopes to reach many more. As with any assistive product, it is being tested carefully in real conditions before any claims of dependable navigation.
What this story shows
For assistive technology, user discovery comes first. The team learned from the people they were building for before deciding what to build.
Frequently asked questions
What is Cue?
Cue is a pair of AI smart glasses connected to a phone that detect objects ahead, recognise text and currency, and read information aloud for visually impaired people.
Who built Cue?
G. Divyasree (BTF-010), K. Rishitha Purna (BTF-011) and Keshamoni Nikhila (BTF-012), third-year CSE-DS students at MVSR Engineering College, in Build Theory Cohort 01.
Is Cue available?
It is being piloted with visually impaired users.
Build your own product in Cohort 02
This story is part of Build Theory Cohort 01: what 24 fellows built. Build Theory is a 30-day online fellowship, based in Hyderabad, where students and early professionals go from a real problem to a live product with 1:1 mentorship and AI tools. Any year, any branch. Check your readiness score or apply for Cohort 02.
- #cue
- #cue smart glasses
- #assistive technology for blind
- #object detection ml model
- #visually impaired navigation
- #social impact startup india
- #g. divyasree
- #k. rishitha purna
- #keshamoni nikhila
- #btf-010
- #btf-011
- #btf-012
- #build theory fellow stories
- #build theory cohort 01
- #student founders india
- #ai product building
- #build theory fellowship