AI Automation & Chatbot Developer | Flutter Developer
AI Chatbot for Customer Support Automation
This AI chatbot helps businesses automate customer queries and reduce response time by up to 60%.
🔹 Problem:
Businesses often struggle with delayed responses, repetitive queries, and high support workload.
🔹 Solution:
Developed an AI chatbot using NLP that can understand user intent and provide instant, accurate replies.
🔹 Key Features:
• Handles common customer queries automatically
• Basic intent recognition using NLP
• Fast and real-time responses
• Easy integration with web/mobile apps
🔹 Result:
• Reduced response time
• Improved customer experience
• Lower support workload
1
510
Created a short AI-based video showcasing how artificial intelligence is transforming the way we work and create.
Worked on concept development, video editing, and storytelling. Also highlighted my experience in AI training and sentiment analysis.
#AI #ArtificialIntelligence #VideoEditing #DataAnalysis #ContentCreation
1
258
I worked on a Sentiment Analysis project where I classified customer reviews into positive, negative, and neutral categories. I performed text preprocessing, used TF-IDF for feature extraction, and applied a machine learning model to improve accuracy. This project helped me gain hands-on experience in NLP, data labeling, and model evaluation.
2
3
382
Built an AI-powered resume analyzer that evaluates resumes and provides instant feedback to improve job success rates.
The system analyzes resume content, structure, and keywords using AI to generate a score and actionable suggestions. It helps users optimize their resumes based on industry standards and ATS (Applicant Tracking System) requirements.
This tool is designed for job seekers and professionals to enhance their resumes and increase their chances of getting shortlisted.
"Open to building similar AI-powered tools for businesses". 🚀
4
273
ML & AI Engineer | Computer Vision | LLM Fine-Tuning & RAG
đź“Ś Project Title
Used Car Price Valuation AI — Machine Learning Resale Estimator
đź’ˇ Overview
The secondhand automotive market faces significant price opacity, making accurate valuation challenging for individual sellers, buyers, and dealership networks.
To solve this problem, I built an end-to-end Machine Learning web application that automates vehicle market price discovery. The application ingests 11 core vehicle specifications — including brand model variant, city location, registration year, driven distance, fuel type, transmission type, owner history, fuel efficiency, engine displacement (CC), and horsepower (BHP) — to deliver instant, reliable market price predictions in Lakhs INR alongside an estimated fair-market value range.
⚡ Key Features & Deliverables
High-Accuracy ML Regression Engine: Powered by Scikit-Learn HistGradientBoostingRegressor trained on thousands of vehicle transactions with custom label encodings for 200+ car models across 11 metropolitan markets.
Real-Time Interactive Valuation Dashboard: Built with Streamlit featuring live input sliders, numerical controls, and instant price calculations.
Interactive Resale Value Range & Gauge Visualization: Integrated dynamic Plotly gauge charts providing a visual market valuation indicator and a condition-based fair market range (+/- 8% variance band).
Figma-Grade Human Design Theme: Designed a custom CSS UI system (Obsidian Slate & Indigo Emerald theme: #0B0F19 background, #6366F1 indigo accents) with Plus Jakarta Sans typography and translucent glassmorphic cards.
🛠️ Tech Stack
Python 3.14 • Scikit-Learn • Streamlit • Plotly • Pandas • NumPy • Joblib • CSS3 Glassmorphism
0
34
This research-based computer vision system automates the digital transcription of analogue gauge readings from video input. By pairing deep learning keypoint detection with geometric perspective correction, the system accurately reads dial values in real time and streams logged metrics directly to cloud-based monitoring tools.
Key Technical Contributions & Features
1. YOLOv8 Keypoint Detection Pipeline: Trained a YOLOv8-pose/keypoint model to precisely identify four critical reference coordinates on the gauge dial—the scale minimum point, scale maximum point, pointer center, and pointer tip. Achieved 93% mAP across a hybrid dataset of 10,000 synthetic gauge images and 2,700 annotated real-world video frames.
2. Parallax & Geometric Distortion Correction: Designed a custom geometric mapping algorithm to overcome 3D parallax errors—where non-coplanar pointers and scale faces skew readings from off-center camera angles. The algorithm computes angular sweeps and normalizes scale orientation for precise angular-to-value conversion.
3. Real-Time Video Analytics Backend: Built an automated video processing pipeline using OpenCV and NumPy to capture live feeds, track pointer displacement continuously, and transcribe visual metrics into structured digital numbers.
4. Automated Cloud Logging: Integrated real-time data exporters to continuously push transcribed gauge values to cloud spreadsheets (Google Sheets / Excel) for continuous tracking and telemetry monitoring.
Tech Stack
Languages & Core: Python, NumPy
Computer Vision & Image Processing: OpenCV
Deep Learning & Machine Learning: YOLOv8 (Keypoint Detection), Convolutional Neural Networks (CNN), Scikit-Learn
Data Integration: Google Sheets API
0
70
SOYFIB is a computer vision platform engineered to automate the detection and classification of soybean leaf diseases. Using a novel, hybrid deep learning architecture, the platform extracts high-dimensional visual signatures from crop imagery to perform zero-shot disease identification and provides interactive tools for researchers to dynamically expand the diagnosis database.
Key Highlights & Engineering Contributions
Hybrid Feature Extraction Pipeline: Built a deep learning vision pipeline combining a custom CNN with a Swin Transformer architecture to extract 512-dimensional visual disease fingerprints from soybean leaf images.
Dynamic Zero-Shot Detection Backend: Implemented a zero-shot diagnosis system by indexing extracted feature embeddings into a FAISS vector database, enabling ultra-fast vector matching and similarity retrieval.
Custom Dataset Preprocessing: Built an end-to-end data pipeline to handle inputs of 225 X 225 X 3 dimensions and parse YOLO-formatted annotations (.txt), optimizing feature extraction accuracy.
Web Platform & Researcher Portal: Developed a Flask web application allowing users to upload geolocated crop images and view real-time diagnostics, paired with an admin portal for researchers to register new disease vectors dynamically.
0
81
EyeSense is an enterprise-grade, real-time AI surveillance system designed for automated visitor monitoring and security tracking. The system replaces manual check-in processes by utilizing computer vision techniques to detect, track, and recognize individuals across multiple concurrent video streams, reducing manual security workloads by approximately 60%.
Key Features & Technical Architecture
Multi-Stream Processing: Engineered a low-latency video pipeline using OpenCV operating at 25 FPS, capable of processing 4 active CCTV feeds concurrently without frame drops or memory leaks.
Face Recognition & Indexing: Implemented identity matching by extracting high-dimensional embeddings using HuggingFace FaceNet models and indexing them into a FAISS vector database. Achieved sub-100ms vector matching across a database of 500+ identity profiles.
Automated Alerting Engine: Designed a real-time event engine that flags unrecognized or unauthorized individuals within 3 seconds, automatically triggering instant security alerts via SMS and email.
Management Dashboard & Containerization: Developed a Flask-based web platform for system administration, access logging, real-time visitor approval, and monitoring. Containerized the entire application stack using Docker for standardized deployment across hardware environments.
0
58
AI Automation Expert | n8n · Make.com · Zapier · OpenAI | 10
New to Contra
AI Automation Expert | n8n · Make.com · Zapier · OpenAI | 10
A real estate business needed an efficient system to collect, enrich, and manage high-quality sales leads from public sources — reducing manual work and enabling seamless CRM integration. We built a targeted lead collection and enrichment system, extracting publicly available data, enhancing each lead with company name, role, and industry details, and preparing everything for direct import into HubSpot and Salesforce.
âś… Industry-specific lead collection using keywords and demographics
âś… Lead enrichment with company name, role, and industry data
âś… HubSpot & Salesforce ready CSV export
âś… Scalable, repeatable lead pipeline
Result: Client significantly improved productivity, cut manual lead tasks, and scaled outreach with organized, enriched data ready for immediate use.
0
203
A brand needed to analyze competitor Instagram performance, identify content trends, and refine their marketing strategy — without manual tracking. We built a semi-automated competitor analysis system using a custom scraper, Google Sheets dashboard, and AI-assisted insights — delivering weekly automated reports with actionable strategy recommendations.
âś… Competitor engagement tracking and content analysis
âś… AI-assisted insights on high-performing content
âś… Real-time Google Sheets dashboard
âś… Weekly automated performance reports
âś… Fully policy-compliant public data sourcing
Result: Client gained clear competitor visibility, improved content strategy, and identified new audience engagement opportunities.
Skills & Tools: Web Scraping, Python, Social Media Marketing, Data Extraction, Instagram, Google Sheets, Automation
0
187
An eCommerce brand needed to streamline and scale their Shopify product listings — eliminating manual uploads and ensuring SEO-friendly, high-converting entries across their entire catalog. We built a custom Shopify scraper and automation system that extracted product details, enriched listings with keyword-optimized content, added branded image overlays, and delivered bulk import-ready CSVs.
âś… Automated product data extraction and enrichment
âś… SEO-optimized titles and descriptions
âś… Branded logo overlays on product images
âś… Bulk import-ready CSV & direct Shopify uploads
âś… Clean, uniform formatting across all listings
Result: Transformed a manual product catalog into a scalable, high-converting, hands-free sales engine.
0
198
A brand needed a streamlined DM campaign system for personalized outreach and lead generation — without violating platform policies or using spammy automation. We built a lightweight, platform-compliant DM outreach system with custom messaging templates, audience targeting using hashtags and keywords, and a semi-automated workflow maintaining a genuine human touch.
âś… Hashtag & keyword-based audience targeting
âś… Personalized messaging templates aligned with brand tone
✅ Semi-automated workflow — human touch maintained
âś… Platform-compliant outreach strategy
Result: Improved response rates, consistent outreach performance, and significant time savings — leading to stronger connections and better conversions.
0
185
Web, API and Android Penetration Tester
Comprehensive validation testing, proof-of-concept verification, and structural remediation audits to ensure complete security patch efficacy.
The Objective:
A development organization required a rigorous follow-up assessment to validate that security patches introduced after an initial audit were correctly implemented, completely neutralized the identified threat vectors, and did not introduce new regression flaws into the environment.
The Approach:
I executed a dedicated verification cycle focused specifically on replicating the original attack patterns using the exact tools and exploits from the initial assessment. Every patch was manually tested using Burp Suite Professional, custom Python exploit scripts, and runtime debuggers to confirm the vulnerability's status and measure remediation success against standard CVSS 3.1 metrics.
From a dedicated retesting scope targeting 20 original security findings, I delivered the following lifecycle results:
Closed (Remediated):Â Successfully verified the resolution of 16 findings, including all critical business logic bypasses and sensitive data exposure pathways, confirming they are no longer exploitable.
Bypassed (Incomplete Fixes):Â Identified 2 instances where a development team used client-side input validation instead of server-side sanitization, allowing the original exploit to still bypass the patch.
New Regression:Â Discovered 2 minor configuration errors introduced by changes made to the access control framework during the patching process.
Documentation Provided:Â Delivered updated Proof-of-Concept (PoC) logs and step-by-step documentation for the outstanding issues to guide the team to a permanent fix.
0
105
Comprehensive external/internal network penetration test, cloud infrastructure review, and vulnerability assessment for a production hosting ecosystem.
The Objective:
A production hosting provider required a full-scope security audit to evaluate their perimeter defense, verify compliance with the CIS Benchmarks and OWASP Top 10 for Cloud, and uncover critical misconfigurations before public deployment.
The Approach:
I executed a hybrid security assessment combining automated infrastructure reconnaissance with deep manual configuration checks. Every vulnerability was manually validated using tools like Nmap, Wireshark, and CloudBrute to eliminate false positives and scored using standard CVSS 3.1 metrics.
From a single comprehensive audit, I uncovered 31 reproducible findings:
3 High Severity:Â Exposed database backups publicly accessible via misconfigured AWS S3 buckets, and unauthenticated remote code execution (RCE) vectors on perimeter network appliances.
6 Medium Severity:Â Overly permissive IAM policies granting excessive root privileges, missing multi-factor authentication (MFA) enforcement on management panels, and outdated SSH protocol versions in use.
10 Low Severity:Â Unrestricted directory listings on internal staging servers, insecure network daemon configurations, and loose firewall rules permitting unnecessary inbound traffic.
12 Informational:Â Internal IP address disclosures in HTTP response headers, missing asset tagging, and verbose system version banners.
0
99
Comprehensive web application penetration test, API endpoint analysis, and vulnerability assessment for a production-facing infrastructure ecosystem.
The Objective:
An enterprise client required a full-scope security assessment to evaluate their external attack surface, verify strict alignment with the OWASP Application Security Verification Standard (ASVS Level 3), and uncover critical server-side and business logic flaws before deploying a major code release.
The Approach:
I executed a hybrid security assessment combining automated reconnaissance with deep manual exploitation across the application layer and its integrated APIs. Every vulnerability was manually validated using Burp Suite Professional, SQLmap, and Postman to eliminate false positives and scored using standard CVSS 3.1 metrics.
From a single comprehensive audit, I uncovered 24 reproducible findings:
2 High Severity:Â Broken Object Level Authorization (BOLA/IDOR) on critical billing endpoints and a SQL injection flaw in the primary authentication pathway.
5 Medium Severity:Â Faulty session management allowing session fixation, missing Rate Limiting on public API routes, and Cross-Site Scripting (XSS) via un-sanitized comment inputs.
7 Low Severity:Â Lax CORS configuration rules, missing secure flag attributes on cookies, and verbose server error disclosures.
10 Informational:Â Exposure of outdated software banners and missing HTTP security headers (such as Content-Security-Policy).
0
123
Comprehensive Android application penetration test, reverse engineering analysis, and vulnerability assessment for a production-ready mobile application.
The Objective:
A mobile application provider required a full-scope security audit to evaluate their client-side attack surface, verify compliance with the OWASP Mobile Application Security Verification Standard (MASVS Level 2), and uncover critical runtime flaws before public deployment.
The Approach:
I executed a hybrid security assessment combining static application security testing (SAST), dynamic instrumentation (DAST), and automated reconnaissance. Every vulnerability was manually validated using tools like Frida, Objection, Jadx-GUI, and Burp Suite to eliminate false positives and scored using standard CVSS 3.1 metrics.
From a single comprehensive audit, I uncovered 19 reproducible findings:
1 High Severity:Â Insecure local data storage exposing unencrypted user credentials and private API keys in the shared preferences file.
4 Medium Severity:Â Broken cryptography implementations, insufficient SSL pinning mechanisms permitting Man-in-the-Middle (MitM) attacks, and unprotected exported Activities allowing unauthorized internal intents.
6 Low Severity:Â Lack of root detection, weak code obfuscation rules, and excessive logging of sensitive debug information via Logcat.
8 Informational:Â Outdated third-party SDK dependencies and missing compiler exploit mitigation configurations.
0
110
Full-Stack Architect & SaaS Founder | Building AI Systems
Software Engineering Job Simulation (JPMorgan Chase & Co.)\
I completed a rigorous software engineering simulation where I worked on real-world engineering tasks. This experience focused on financial data visualization, system architecture, and implementing technical solutions to meet industry-standard requirements. It refined my ability to write clean, maintainable code for complex enterprise environments.
Key Contributions:
Data Visualization: Developed tools to visualize live financial data for better decision-making.
System Design: Analyzed and implemented fixes for existing codebases to improve performance.
Industry Standards: Applied professional software development life cycle (SDLC) practices.
Skills: Python, Technical Analysis, Data Visualization, Git.
1
109
Academic Web Portal Modernization (DAVV)
I completed an institutional internship focusing on overhauling and upgrading the official web platform for the School of Life Sciences at DAVV. The goal was to transform a legacy system into a responsive, fast, and user-friendly portal for academic content delivery. My work ensured better layout hierarchy, clean frontend structure, and seamless navigation for high-volume user traffic.
1
61
GetYourQR: AI-Powered QR Menu & Smart Billing SaaS
I built and launched this full-stack platform to modernize the hospitality industry. My focus is on high-efficiency development and using modern frameworks to deliver scalable, production-ready systems with speed and precision.
Key Features:
Dynamic QR menu generation with real-time updates.
Automated billing and order management system.
High-performance cross-platform mobile experience.
2
2
190
OnlyTurf: Integrated Sports Booking & Venue Management System
I developed a dual-platform ecosystem designed to streamline sports facility bookings. This project includes a high-performance mobile app for users to discover and book slots, and a comprehensive web dashboard for vendors to manage their venues, schedules, and payments.
Key Features:
Dual-Interface: Specialized flows for both sports enthusiasts and facility owners.
Real-time Booking: Instant slot reservation with live availability updates.
Vendor Management: Robust administrative tools for slot scheduling and revenue tracking.
1
83
AI/LLM Developer | Python & FastAPI | Custom AI Web App
Turning ideas into simple, useful web experiences
Turning ideas into simple, useful web experiences
AI Automation & Workflow Solutions
AI Support Email Automation & Model Testing
Tested AI models for handling high-volume customer support emails and automatically classifying customer issues.
The work compares models based on:
• Classification results
• Response stability
• Processing latency
• Estimated monthly cost
The testing also included a real-world support email containing multiple issues, such as a payment problem, duplicate charge, missing order confirmation, and app crash.
The goal is to identify an AI approach that can support reliable email classification and automated routing at scale.
Skills:
AI Automation
Artificial Intelligence
Data Analysis
Tools:
Only select the tools you actually used for this work, such as:
Python
Google Colab
OpenAI
0
30
AI Support Email Automation & Model Testing
Tested AI models for handling high-volume customer support emails and automatically classifying customer issues.
The work compares models based on:
• Classification results
• Response stability
• Processing latency
• Estimated monthly cost
The testing also included a real-world support email containing multiple issues, such as a payment problem, duplicate charge, missing order confirmation, and app crash.
The goal is to identify an AI approach that can support reliable email classification and automated routing at scale.
0
60
AI Business Automation
A practical AI and workflow automation solution designed to reduce repetitive work and connect business processes.
This work covers lead automation, CRM integration, AI assistants, data automation, reporting, customer support workflows, and tool integrations.
The approach is simple: understand the existing process, identify what can be automated, build the workflow, connect the required tools, and test the solution.
0
61
Lead & CRM Workflow Automation
A business workflow concept designed to automate the process from lead capture to CRM updates and follow-up.
The workflow connects incoming leads with automated data processing, lead qualification, CRM updates, personalized follow-ups, and internal notifications.
The goal is to reduce repetitive manual work, keep customer information organized, and ensure leads are followed up consistently.
0
31