H
Best Idea Award Winner — Composio Agents in Production Hackathon

Horizon

AI Recruiter Copilot

An intelligent recruitment automation pipeline built with LangGraph & Composio. From monitoring Gmail to scoring candidates and scheduling interviews.

System Architecture

Powered by LangGraph, the pipeline executes an 8-node state machine with conditional routing, orchestrating 4 Composio toolsets and Groq LLM inference.

1. LangGraph State Machine

1. Gmail-Monitor

Auto-downloads incoming resumes

2. .pdf extractor

PDF/TXT parsed to structured data

3. LinkedIn + LLM Enricher

LinkedIn API + Groq LLM fallback

4. Profile Scorer

Groq evaluates vs job criteria

Recruitment Agent

Evaluates condition: Score > 6.5

✅ Score > 6.5 (Shortlist)

6. Scheduler

Creates Google Calendar events

8. G Sheet 2 & G Calendar

Scheduled Interview output

❌ Score <= 6.5 (Reject)

7. G Sheet 1 (All Candidates)

Export to Database / JSON

* Skips scheduling, goes directly to DB export.

2. Detailed Systems Diagram

Interactive system design. You can pan, zoom, and inspect the different components.

Loading interactive diagram...

Pipeline Execution

Watch the automated recruitment pipeline run in real-time, executing the complete process from email reading to scheduling.

powershell - python ai_recruiter_pipeline.py

Click "Run Demo" to start the pipeline simulation

Automated Outputs

Master Candidate Database

Google Sheets Integration

Waiting for pipeline execution...

Interview Scheduling

Google Calendar Integration

Waiting for pipeline execution...

-
Shortlisted
-
Rejected

Powered by Composio & Groq

Horizon connects directly to real-world applications using Composio's authenticated toolsets, ensuring secure and reliable automation.

Gmail

Automated Resume Monitoring

LinkedIn

API Data Enrichment

Google Calendar

Interview Event Scheduling

Google Sheets

Structured Master Database

Groq Llama 3.1

High-speed AI Scoring & Parsing

LangGraph

State Machine Orchestration