QuantumAI

Your personal guide to Prabhat Kumar's portfolio.

Ask me anything about Prabhat's skills, experience, or projects.
AI OutreachSpring BootLead AutomationB2B SaaS

AI-Powered Sales Outreach & Lead Automation Platform — Architecture & Engineering

This platform turns outbound sales into an automated engineering pipeline: discover leads, validate contacts, personalize messages, run campaigns, classify replies, and surface real-time analytics.

24/7 Laptop-Hosted Automation System

The system runs continuously on Prabhat's laptop and coordinates lead discovery, email validation, campaign execution, AI personalization, automated follow-up sequences, reply classification, and dashboard analytics.

It is designed for B2B lead generation, SaaS outbound sales, and agency prospecting where personalization and response handling matter more than raw volume.

LinkedIn, Google Maps, and company website discovery

MX, SMTP, and domain reputation validation

OpenAI-powered personalization

Interested, Not Interested, Schedule Later, and Demo Request classification

Microservice-Oriented Stack

The backend uses Java, Spring Boot, Spring Security, JWT, PostgreSQL, Redis, and Kafka or RabbitMQ for asynchronous workflows. Python workers handle scraping, enrichment, and AI-heavy tasks where the ecosystem is stronger.

React and Next.js power the dashboard, Docker packages the stack, and Redis keeps rate limits, campaign state, and temporary enrichment results fast.

Lead Sources -> Discovery Workers -> Validation Queue -> CRM Store
CRM Store -> AI Personalization -> Campaign Manager -> Follow-up Engine
Inbox Webhooks -> AI Reply Classifier -> Auto Reply + Workflow Automation
Events -> Analytics Stream -> Realtime Dashboard

Reply Classification and Workflow Automation

Inbound replies are classified into business states instead of being dumped into an inbox. That makes it possible to auto-send meeting links, stop campaigns for uninterested leads, or schedule later follow-ups without manual sorting.

Examplejava
public enum ReplyIntent {    INTERESTED,    NOT_INTERESTED,    SCHEDULE_LATER,    DEMO_REQUEST}public void handleReply(LeadReply reply) {    ReplyIntent intent = aiClassifier.classify(reply);    workflowRouter.route(reply.leadId(), intent);}

Work with Prabhat

Build production-grade backend systems, AI workflows, cloud automation, and high-signal engineering products with a developer who ships from architecture to deployment.

Work with PrabhatContact