DIVYANSHI SINGH
@Divyanshi018572AI/ML Engineer | Building intelligent systems with LLMs, RAG Systems, Multi-Agent AI & Deep Learning | Open Source Enthusiast
Language Breakdown
Lines of code distribution across 28 owned repositories
I-Shaped Developer
I-shapedSpecialist — deep expertise in Jupyter Notebook
Collaboration Network
Global Impact visualization
Repos
33
PRs
0
Growth
+18%
Top Collaborators
No collaborator data yet.
Coding Streak
Contribution activity over the past year
Michael Wang
@DrMichaelWang
KuAnjalii
@KuAnjalii
𝐎𝐬𝐚𝐬𝐚𝐧 𝐎𝐥𝐮𝐬𝐨𝐥𝐚
@incognito-dev07
Daniel Cela
@CelaDaniel
Braedon Saunders
@braedonsaunders
Top Repositories
An AI-powered Road Damage Assessment system using YOLOv11s & SAHI to detect potholes and cracks with high precision. Featuring a Streamlit dashboard for real-time analysis, it provides engineering reports based on IRC standards. Integrated with RAG technology to offer expert maintenance consulting derived from civil engineering manuals.
Business Intelligence dashboard analyzing 400K+ Meta ad events using Power BI, Excel & DAX.
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
Invoice Scanner Pro automates receipt processing using EasyOCR and Streamlit. It instantly extracts key financial data, allowing manual correction, SQLite storage, and real-time CSV exports.
ResumeIntel Nexus is a specialized AI application for deep single-resume intelligence. Powered by Groq and Llama 3.3, it offers interactive skills gap heatmaps, actionable improvement feedback, and a contextual chat sidecar. Bridge the gap between JD requirements and profiles with local persistence and high-performance analysis.
OllamaRAG Nexus is a professional-grade suite that leverages local Ollama models and LangChain to provide secure, high-speed document intelligence. Featuring a dynamic RAG pipeline, it allows users to upload PDF/TXT files into a persistent ChromaDB vector store for instant, context-aware chatting. Experience premium, private AI locally.
Movie Match Maker is a full-stack, AI-powered recommendation platform built with FastAPI and Streamlit. It uses an NLP/TF-IDF machine learning engine and Groq LLMs to provide personalized movie suggestions via natural language search. Integrated with the TMDB API, it features a glassmorphism UI, rich media assets, and a smart watchlist analyzer.
An applied Machine Learning project utilizing XGBoost to solve core supply chain inefficiencies. By applying frequency encoding to geographic data, this ML system achieves 83% accuracy in predicting late delivery bottlenecks. Concurrently, an imbalanced learning pipeline (SMOTE) captures 97% of fraudulent transactions for manual review.
Loan Default Prediction is an end-to-end fintech risk analytics project built on the Lending Club dataset (accepted loans from 2007-2018). The objective is to estimate the probability of borrower default and convert model output into practical underwriting decisions: Auto Approve, Manual Review, or Decline.
An applied ML project for fraud detection where accuracy is deprioritized in favor of PR-driven metrics. Integrates Isolation Forest, One-Class SVM, Random Forest, and XGBoost with SMOTE and threshold tuning, plus a decision-focused UI for analysts to test fraud scenarios interactively.
Open Source Impact
Contributions to external projects
No external contributions found.