
Building AI tools for vision, language & automation. Exploring offline LLMs, Re-ID, and intelligent pipelines.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Court-Case-Metadata-Fetcher-Flask-Selenium-UI-
August 2, 2025 – August 2, 2025
A Flask web app to fetch Indian court case details using Case Type, Number, and Year with simulated output due to CAPTCHA restrictions.
View Projectdefi-credit-scoring
July 24, 2025 – July 24, 2025
A machine learning-based credit scoring engine for DeFi wallets on the Aave V2 protocol, using on-chain transaction history to generate scores between 0–1000.
View Projectrecommendation_engine
July 23, 2025 – July 23, 2025
A lightweight AI-powered astrologer recommendation system built using semantic similarity and local LLMs. The project includes a semantic matching engine for astrologer suggestions based on user queries and an AI astrologer agent that generates personalized astrological insights using the Phi-3 model via Ollama. Implemented in Python
View Projectairline-demand-insights
July 18, 2025 – July 18, 2025
Airline Demand Trends Analyzer A data-driven web application that visualizes market demand trends in the airline industry using real-world flight booking data. The system analyzes a large dataset of global flights to identify top origin countries and high-demand routes, offering valuable insights for travel companies, marketers.
View ProjectDetection-Summary-Engine
July 17, 2025 – July 17, 2025
Detection Summary Engine – A real-time object detection and summarization system that processes video feeds, detects key entities (e.g., players, balls, objects), and generates structured event summaries. Designed for sports analytics, surveillance, and automated monitoring using deep learning and OpenCV.
View ProjectQuery-Based-Image-Retrieval-System
July 17, 2025 – July 17, 2025
Query-Based Image Retrieval System (Person Re-ID) – A computer vision pipeline for clustering and retrieving images of the same person across varied scenes using face and clothing features. Combines InsightFace, ResNet-50, FAISS, and DBSCAN to achieve accurate, unsupervised person re-identification from visual queries. Ideal for surveillance
View ProjectLinkedIn-LLM-Post-Generator
July 17, 2025 – July 17, 2025
LinkedIn LLM Post Generator – Generate high-quality, professional LinkedIn posts completely offline using a fine-tuned open-source model (TinyLLaMA-1.1B-Chat). Built with PEFT + LoRA on Apple M3, this project requires no cloud APIs and supports themes like internships, promotions, certifications, and project completions.
View ProjectVision-Pivot
July 17, 2025 – July 17, 2025
Vision Pivot – A real-time, face-based attendance system powered by Python (FaceNet, OpenCV) and a React + Firebase admin dashboard. Secure, scalable, and instantly synced for modern organizations.
View ProjectReddit-user-persona
July 14, 2025 – July 14, 2025
Reddit User Persona – Generate personality profiles and behavioral summaries from Reddit usernames using local LLMs like Ollama. Analyze posts & comments to uncover user traits, interests, and tone — all with full privacy, no external API calls.
View ProjectCultural Fit Analysis
The candidate demonstrates a strong passion for data science and AI through a diverse portfolio of personal projects. The projects cover various domains (NLP, computer vision, finance, analytics), indicating adaptability and a broad interest in applying AI solutions. The focus on building practical, deployable systems aligns well with a results-oriented culture. However, the lack of team-based projects or professional experience makes it challenging to fully assess collaborative cultural fit.
Soft Skills & Operational Fit
The candidate's project descriptions indicate a strong drive for independent project development and problem-solving. The focus on offline and privacy-preserving solutions (e.g., local LLMs) suggests an innovative and resourceful approach. However, without psychometric or English test scores, it's difficult to assess communication clarity, teamwork, or stress handling abilities.