Product story
AI Engineering · 2 min read
Capturing the context behind every contact
Extracting 15 to 28 fields from each card, then capturing where you met and why while the conversation is still fresh.
- Client
- Tejvon product: AI Lead Capture
- Industry
- Sales technology
- Capabilities
- AI Engineering, Mobile Apps
- Status
- Beta
Highlights
- 15 to 28 fields extracted per card
- Review time spent only on low-confidence fields
The challenge
A stack of cards a week after an expo is a list of names with no memory attached. The printed fields say who someone is, not why you kept their card.
Our solution
- 01Extracted every printed field: name, designation, company, services offered, mobile, WhatsApp, LinkedIn and other social profiles.
- 02Added two fields no card prints: where you met and the purpose, captured in seconds during the scan.
- 03Showed a confidence percentage on every field, so users check only the uncertain ones.
- 04Let users correct any field or add what the card did not have before saving.
What we delivered
- Extraction of 15 to 28 fields per card
- Context capture for meeting place and purpose
- Review screen with per-field confidence
Impact
15 to 28fields extracted per card
Review time spent only on low-confidence fields
Technology
KotlinJetpack ComposeML KitRoom
Have a problem like this?
Start with a free 1-hour architecture review with a senior engineer.