✦ Vendor discovery · Algolia · AWS · Expo

Klik — Vendor Discovery Platform

Ongoing

Klik is a vendor discovery and directory platform built for UK and Nigerian diaspora markets — helping people find trusted service businesses in hair, beauty, events, and media with structured profiles and real search, instead of relying on Instagram DMs and WhatsApp recommendations.

View site

Stack: ReactNext.jsNode.jsAWS AmplifyDynamoDBS3API GatewayAlgoliaAlgolia RecommendNextAuth.jsAWS CognitoPythonExpoReact NativeR&D prototype: FAISS, OpenCLIP, LLM integration, Instagram scraping
Status: Ongoing
Year: —
Website: www.klikonline.co.uk

OVERVIEW

THE PROBLEM

In these markets, discovery runs almost entirely through informal networks. The friction isn't just finding a name — it's trusting fit, quality, and responsiveness, and then getting from interest to a confirmed booking without endless back-and-forth across multiple platforms. There's no Trustpilot for a Lagos hair stylist. No verified profile for a London-based Nigerian makeup artist. Just screenshots, referrals, and hope.

WHAT WAS BUILT

✦Full web platform — vendor signup, authentication, onboarding flow, profile management, and vendor dashboard
✦Customer-facing discovery — search and browse experience powered by Algolia, with filters, pagination, and related vendor recommendations via Algolia Recommend
✦Platform architecture — Next.js frontend, AWS Amplify Lambdas, DynamoDB, S3, API Gateway, NextAuth and AWS Cognito for vendor authentication and sessions
✦Mobile app — Expo (React Native) client in active development, reusing Algolia and vendor APIs for on-the-go discovery
✦Explorations (separate R&D) — As a standalone MVP and research prototype: a multimodal retrieval system using FAISS, OpenCLIP (ViT-B-16), and Instagram-scraped vendor data — exploring AI-powered vendor matching as a future direction for the platform. Not currently powering klikonline.co.uk.

THE INSIGHT

The real lesson from Klik had nothing to do with the technology. User research revealed that informal, relationship-based booking isn't a problem to be solved — it's how these markets deliberately operate. Vendors manage availability through personal relationships, not calendars. Engineering around that wasn't going to work. Knowing when to stop building is a product skill. Klik is where I learned it.

STACK & TOOLS