PN: 2483-AX9
DO NOT REMOVE DURING OPERATION
SN: 002194-C
MAT: AL6061-T6
信頼
Book a call
Client
BacklinksAI

AI-powered media opportunity matching for HARO

Impact of
our work

プロジェクトについて
Project type
AI Media Matching Platform
Tech stack
Next.js
Prisma
PostgreSQL
Pinecone
OpenAI
built in
3+ years & ongoing
AI matching
profile-to-prompt matching
Notifications
email digests + Slack alerts
AI Autopilot
auto-drafted responses

About
the project

プロジェクト
BacklinksAI is an AI-powered platform that helps users find and match with HARO (Help A Reporter Out) media opportunities. Journalists use HARO to request expert sources for their stories. BacklinksAI uses AI and vector embeddings to intelligently match user profiles with relevant media queries based on their expertise, company information, and background.
The platform automates the discovery of backlink and PR opportunities by analyzing incoming HARO requests against user profiles, scoring confidence levels, and delivering matched opportunities through email and Slack notifications.

The
challenge

挑戦
ID.5.001
Build an intelligent matching system that goes beyond keyword search to understand the nuance of journalist requests and user expertise, then deliver timely matches across multiple notification channels.

The
solution

解決策
ID.1.001
Vector-based semantic matching
Pinecone vector database with OpenAI embeddings for intelligent prompt-to-profile matching with confidence scoring.
ID.1.002
Background job pipeline
BullMQ + Redis for HARO email ingestion, prompt parsing, AI matching, and scheduled email delivery.
ID.1.003
AI writing assistant
Context-aware response drafting using OpenAI, customizable to each user's writing style and expertise.
ID.1.004
Firecrawl integration
Automated web scraping of user company websites to enrich profile data for better matching.
ID.1.005
Separate email worker service
Dedicated Node.js service with BullMQ job queue for reliable, scheduled email delivery using React Email templates.
ID.1.000
BYLD built BacklinksAI from the ground up as a full-stack platform. The core matching engine uses OpenAI embeddings stored in Pinecone's vector database to perform semantic search, comparing user profiles against incoming HARO prompts and scoring confidence from 0-100%.
We built a background job system with BullMQ and Redis to handle HARO email ingestion, prompt parsing, vector matching, and scheduled notification delivery. Users receive matched opportunities through customizable email digests powered by React Email and SendGrid, or instantly via Slack integration.
The platform includes a multi-step onboarding flow, Stripe subscription management with free and standard plans, and an AI writing assistant ("Jarvis") that helps users compose responses to matched opportunities.

Impact of
our work

プロジェクトについて
AI matching
profile-to-prompt matching
Notifications
email digests + Slack alerts
AI Autopilot
auto-drafted responses