AI · B2B sales
Bizigniter
Swipe-style B2B prospecting, guided by AI
Year
2025 — prezent
Our role
Our own product — AI engine, backend, web and mobile apps.
Status
in production
0–100
AI fit score
Web + Mobile
shared monorepo
pgvector
semantic recommendations
The challenge
Manual prospecting is slow and tedious — stale lists, no financial context, no prioritization. Bizigniter had to make client discovery as simple as a swipe, and learn from every decision the user makes.
The approach
A recommendation engine (Python / pgvector) proposes companies scored 0–100 with explanations; the app hydrates them with balance sheets and risk signals from Companero. A NestJS backend, a Next.js web app and an Expo mobile app from a single monorepo with shared types. Every swipe feeds back into the engine, which gets steadily more personalized — with a graceful fallback when the AI is unavailable.
What stands out
- Daily swipe-style discovery, with AI-scored companies and "why we recommend this" reasons
- 5-year financial health and risk signals on every card
- A mini-CRM with a staged pipeline, tags and notes
- Feedback loop: every swipe personalizes future recommendations
- Web (Next.js) and mobile (Expo) from one monorepo with shared types
Tech stack
Next project
Companero