Rana shahmeer ali
Sr Software Engineer With 4+ Yrs Exp in Product Development
New to Contra
Rana is ready for their next project!
📈 Real-Time Stock Data Streaming Pipeline | AWS, Kafka & Python
Built a real-time stock data streaming pipeline using Python, Apache Kafka, and AWS to process live market data with low latency. The solution streams stock data into Kafka, stores it in Amazon S3, and enables real-time analytics through AWS Glue, replacing slow batch-based ETL workflows.
The architecture reduced data processing latency from hours to seconds, providing a scalable and efficient streaming data pipeline.
🛠️ Tech Stack
Language: Python
Streaming: Apache Kafka
Cloud: AWS Lambda, Amazon S3, AWS Glue
Automation: Python ETL
🌟 Key Features
📡 Real-time stock data ingestion
⚡ Kafka-based streaming pipeline
☁️ Automated ETL with AWS
🗄️ Data lake integration using Amazon S3
📊 Scalable real-time analytics
🚀 Reduced ETL latency from hours to seconds
0
18
🤖 AI Lead Enrichment & Cold Email Automation | n8n, OpenAI & Google Sheets
Proud to showcase an AI-powered lead enrichment and cold email automation system built entirely in n8n. This workflow transforms raw prospect data into highly personalized outreach campaigns by enriching company information, intelligently scoring leads, generating AI-written cold emails, and automatically sending them at scale—all while keeping costs under $0.001 per lead.
Here's a look behind the workflow, architecture, and automation that powers this intelligent outbound sales pipeline. 👇
🛠️ Tech Stack
Automation Platform: n8n
AI Model: OpenAI GPT-4o-mini
Data Source: Google Sheets
Email Service: Gmail API
Lead Enrichment: Third-party Company & Contact APIs
Workflow Logic: Rule-Based Lead Scoring Engine
Monitoring: Real-time logging & error handling
🌟 Key Features Developed
1. 🎯 Intelligent Lead Enrichment
Automatically imports raw leads from Google Sheets.
Enriches company and contact information using external APIs.
Handles missing or incomplete enrichment data with graceful fallbacks.
Filters invalid or unusable email addresses before outreach begins.
2. 🧠 AI-Powered Lead Scoring
Custom 0–100 lead scoring engine based on 8 qualification signals.
Prioritizes high-value prospects for better conversion rates.
Rule-based evaluation ensures transparent and consistent scoring.
Easily configurable scoring criteria for different business needs.
3. ✉️ Personalized Cold Email Generation
Generates unique, personalized cold emails for every lead using OpenAI GPT-4o-mini.
Tailors messaging using enriched company information.
Eliminates repetitive templates and improves engagement.
Produces human-like outreach while maintaining brand consistency.
4. ⚡ Automated Outreach & Delivery
Sends emails automatically through Gmail.
Built-in rate limiting to comply with sending limits.
Duplicate email prevention ensures each lead is contacted only once.
Real-time delivery tracking for every outreach attempt.
5. 📊 Workflow Monitoring & Reporting
Logs every outcome (Sent, Skipped, or Failed) back to Google Sheets.
Centralized workflow monitoring with global error handling.
Detailed execution history for troubleshooting and reporting.
Reliable retry and failure management for uninterrupted automation.
🚀 Highlights
🤖 Fully automated outbound sales workflow
📊 AI-powered lead qualification
✉️ Personalized email generation with OpenAI
⚡ End-to-end workflow built in n8n
📈 Real-time Google Sheets synchronization
🚫 Duplicate send prevention
✅ Invalid email filtering
🛡️ Robust error handling & fallback mechanisms
💰 End-to-end processing cost of less than $0.001 per lead
🧩 Technologies Used
n8n
OpenAI GPT-4o-mini
Google Sheets
Gmail API
REST APIs
AI Automation
Workflow Automation
Lead Enrichment
Lead Scoring
Cold Email Automation
Business Process Automation
0
17
🚀 Full-Stack Multi-Vendor E-Commerce Marketplace | Next.js, FastAPI & PostgreSQL
Proud to showcase My App, a next-generation, high-performance multi-vendor e-commerce marketplace built from the ground up to deliver a seamless shopping experience for customers, a feature-rich portal for vendors, and a premium management dashboard for administrators.
Here is a look behind the scenes at the tech stack, architecture, and features that power this modern marketplace! 👇
🛠️ The Tech Stack
Frontend (Customer App & Admin Panel): Next.js (React), TypeScript, Tailwind CSS
Backend API: FastAPI (Python), SQLModel, Uvicorn
Database: PostgreSQL (Relational schema designed for scale)
Integrations: PayFast, JazzCash, EasyPaisa (Multi-channel payment gateways)
SEO & Crawling: Dynamically generated XML sitemaps, semantic HTML, and custom crawlers configuration
🌟 Key Features Developed
1. 🛍️ Customer Experience (Next.js)
Dynamic Catalog & Advanced Filters: Lightning-fast searching, category navigation, and attribute-based sorting.
Optimized Product Detail Pages (PDP): Detailed specification grids, dynamic pricing adjustments, and full content metrics formatter.
Instant SEO Routing: Real-time generation of sitemap.xml and robots.txt ensuring automatic indexing of thousands of products by Google.
2. 💼 Vendor & Admin Control Rooms
Unified Dashboard: Live reports on sales statistics, commissions, returns, and order fulfillment progress.
Inventory & Import Management: Multi-attribute product configuration (Simple and Variable models) with bulk Excel import validation and automatic SKU generators.
Pricing Safety Guardrails: Built-in validation blocking zero-price or invalid sale-price listings to protect store margins.
3. ⚡ High-Performance FastAPI Backend
Type-Safe REST APIs: Built using FastAPI and SQLModel for fast execution speeds, automatic documentation, and secure schema validations.
Transaction Loggers: Tracking store inventory additions, price changes, and customer transactions.
1
1
39
Automated Data Lineage & Native SQL AST Engine
Tools / Tech Stack (Tags):
Data Engineering SQL Server (T-SQL) ETL Automation Data Governance Dynamic SQL C# SQLCLR
The Problem:
In complex regulatory reporting systems, determining why a specific financial account ended up in a specific reporting cell is a massive manual headache. The system stored high-level rules, but the underlying business logic was buried deep inside layers of SQL Views, bit-flags, and General Ledger (GL) mapping tables. When business stakeholders needed answers, developers had to spend hours manually reverse-engineering SQL code and translating raw database IDs into human-readable reasons.
The Solution:
I architected and deployed a 100% automated, zero-dependency Data Lineage Engine natively within SQL Server. This engine acts like an automated developer—it dynamically reads the SQL Views, unpacks the hidden business logic on the fly, cross-references live account data, and translates everything into plain English.
My Contributions & Deliverables:
Native T-SQL AST Parser: Engineered a highly optimized string parsing engine in pure T-SQL. It dynamically reads SQL View definitions, mathematically tracks nested parentheses, and extracts embedded CASE WHEN logic—completely bypassing the need for external Python/CLR dependencies or infrastructure changes.
Multi-Layer Logic Unpacking: Designed an automated ETL pipeline that parses high-level metadata constraints, resolves deep-level bit-flag dependencies, and maps them to dynamic GL account arrays.
Automated Data Translation Engine: Built a dynamic resolution layer that converts raw database IDs and foreign keys (e.g., CountryID = 77) into human-readable business logic (e.g., 'United Kingdom') using a centralized dictionary mapping schema.
Master Orchestration: Delivered a foolproof, idempotent batch processing architecture that loops through thousands of reporting cells and populates a finalized lineage matrix for stakeholders in seconds.
The Impact:
Transformed a manual, hours-long investigative task into a fully automated, one-click pipeline. Business users now have instant access to a transparent, human-readable lineage matrix that guarantees 100% accurate data governance and regulatory compliance reporting—with zero external server dependencies.
0
59