Agentic SystemsSTATUS: Open Source 1 0

AI Sales Lead Qualifier

Automated intent scoring and routing agent graph

Built with Python, LangGraph, and FastAPI, this system serves as intelligent middleware between inbound inquiry forms and CRM platforms. It parses unstructured lead communication, extracts key budget and decision-maker entities, scores intent, and updates CRM records.

Inspect GitHub Source Code

01 // THE BUSINESS PROBLEM & SITUATION

Operational Context

Sales teams often spend significant manual effort reviewing unqualified inbound web inquiries, delaying response times for high-priority prospects.

CLIENT / ENVIRONMENT: Designed as a scalable solution for B2B sales teams dealing with high-volume inbound marketing traffic.

02 // SYSTEM ARCHITECTURE & SOLUTION

What Was Built

I architected an agentic workflow using LangGraph that intercepts incoming webhooks, analyzes intent against a qualification rubric, and interacts with target CRM APIs.

SALES.QUALIFIER.AI // SUPERVISOR_GRAPHConceptual Architecture Diagram

INPUT

Web Inquiry

ROUTER

LangGraph

EVALUATION

Intent Score

OUTPUT

CRM Record

SYSTEMS_ARCHITECTURE_EXPLODER // LAYERED_STACKFULL-STACK SYSTEM DECOUPLING

Frontend Layer

Next.js 16 App Router, React 19, Tailwind CSS, TypeScript

API & Middleware

FastAPI, Pydantic Schema Validation, Rate Limiters

Orchestration

LangGraph Multi-Agent Supervisor & State Machines

Model Providers

OpenAI GPT-4o, Anthropic Claude 3.5, Vision AI

Data Stores

PostgreSQL (Prisma), Qdrant Vector Store, Redis Caching

Observability

LangSmith traces, Pydantic audit logs, Error retries

03 // ENGINEERING CONTRIBUTION & FEATURES

Exact Contribution

Architected the multi-agent supervisor graph in LangGraph
Developed the FastAPI backend middleware with Pydantic validation
Integrated CRM API webhooks with retry queueing
Designed the administrative monitoring interface

Key Capabilities Built

✓ Context-aware Intent Scoring

✓ Automated CRM Entity Extraction

✓ Real-time Representative Alerting

✓ Structured Pydantic Output Validation

DATA_PROVENANCE_TIMELINE // AUDIT_TRACETRACE_ID: #TR-992041

14:22:01.002Source Ingestion

Received webhook payloadOK

14:22:01.045Schema Validation

Pydantic structured field checkOK

14:22:01.210LangGraph Reasoning

Evaluated criteria rulesOK

14:22:01.350Human Audit Check

Confidence threshold evaluationPASSED

14:22:01.480Database Commit

Created audit record in PostgreSQLCOMMITTED

*Illustrative system trace — demonstration data, not a client result.

04 // EXECUTION PIPELINE

Step-by-Step Workflow

STEP 01Prospect submits an inquiry form.

STEP 02FastAPI receives the webhook payload.

STEP 03LangGraph agent extracts buying signals and evaluates criteria.

STEP 04System validates JSON schema before external dispatch.

STEP 05Qualified prospects trigger immediate representative notification.

05 // ENGINEERING LESSONS & OUTCOMES

Technical Challenges Overcome

• Mitigating model hallucinations during criteria scoring.
• Handling upstream API rate limits under sudden traffic spikes.

Measurable Outcomes

→ Reduced initial qualification response time in testing.

→ Automated initial evaluation for low-intent inquiry patterns.

→ Provided structured JSON schemas for CRM ingestion.

Lessons & Engineering Rules

"Model outputs should be validated against schemas before external API dispatches."
"Isolating agent reasoning steps from action execution helps prevent partial state corruption."

NEXT RELATED CASE STUDY

OmniRouter Agent Dispatcher

View Study

PROJECT METADATA

ARSLAN'S ROLEAI Engineer & Systems Architect

YEAR & STATUS2025 // Open Source

CATEGORYAgentic Systems

TECHNOLOGY STACK

PythonLangGraphFastAPINext.jsPostgreSQLPydanticDocker

GITHUB METRICS

Stars: 1

Forks: 0

Verified GitHub Update: 8/2/2026

Need a similar solution?

If your business faces similar data, automation, or software bottlenecks, let's discuss your requirements.

Start a Conversation