What I Built
I built VEYRA (New Era of Viewing and Managing Emails) around a common problem I saw with a friend: managing multiple inboxes and constantly checking which emails actually need attention.
VEYRA is an AI-powered email intelligence dashboard that turns a crowded inbox into an Attention Center. Instead of simply showing unread counts, it identifies what is important, what requires action, what has a deadline, and what can wait.
The goal is simple: help people spend less time deciding which emails matter.
Demo
Code
How I Built It
VEYRA is built with Python, Flask, HTML, CSS and Vanilla JavaScript.
For AI intelligence, I integrated the open-weight Qwen/Qwen2.5-1.5B-Instruct model through the Hugging Face Inference API.
The model is used for email analysis such as summaries, categories, priorities, actions and deadlines.
I also built a local deterministic fallback so the application continues working when the external AI service is unavailable.
This makes the AI layer replaceable while keeping the core product functional.
Why Does Open Innovation Matter?
Open innovation made it possible for me to experiment with an AI model that I can actually build around and control instead of treating AI as a black-box feature.
For a project like VEYRA, this matters because email intelligence needs to be adaptable. I can change the model, experiment with different open-weight models, add local processing, or improve the fallback without rebuilding the entire application around one closed provider.
It also makes the architecture more transparent and gives the project a path toward more privacy-conscious, user-controlled email intelligence.
Prize Categories
- Best Use of Render — VEYRA is deployed and publicly accessible through Render.
Top comments (0)