Freelancers using Streamlit in ThaneFreelancers using Streamlit in Thane
I build AI agents & internal tools that act on data
    13
    Followers
I build AI agents & internal tools that act on data
I’m an AI & Machine Learning engineer with expertise in deve
    6
    Followers
I’m an AI & Machine Learning engineer with expertise in deve
Cover image for It all started on a
It all started on a Sunday at the AWS User Group Mumbai meetup. I wasn't expecting to walk away with a new obsession, but then the speaker introduced me to Temporal and everything changed. Temporal is a durable execution engine that solves one of the hardest problems in agentic AI what happens when your LLM workflow crashes mid-run? Normally you lose everything So I went home and built this: an agent that monitors your competitors around the clock tracking pricing changes, product launches, hiring signals, and strategic moves. Every 24 hours it uses Mistral (running fully on-device via Ollama) to analyze the data and synthesize a structured executive briefing delivered straight to your inbox. Sometimes the best projects start with a Sunday conversation. https://github.com/AnuragNagare/Agentic-AI-.git
0
102
Cover image for launched TextGenix Enterprise — an AI-powered intelligent do...
launched TextGenix Enterprise — an AI-powered intelligent document processing system! This platform enhances and transforms documents (PDF, DOCX, TXT, HTML, RTF) with context-aware vocabulary improvements, grammar validation, and industry-specific terminology (legal, medical, financial, technical). It comes with a sleek Gradio-based web interface featuring modern styling, interactive analytics dashboards, and real-time quality metrics like semantic preservation, grammar score, and AI confidence levels. If you’re looking to build your own AI-powered text/document platform, enhance business workflows with custom NLP models, or integrate analytics-driven AI solutions into your enterprise apps I can help.
0
123
Cover image for Neuro Lend
My credit model scored
Neuro Lend My credit model scored 0.92 AUC. Then I changed one field on one applicant, female to male, and she went from approved to rejected. A lot of lending AI works like this: it says no in milliseconds, and nobody, not the customer and often not the bank, can say why. So I built NeuroLend. You fill in a loan application and watch the signal travel through every neuron until the output turns green or red. Then it shows why it decided, and the smallest change that would flip the answer. Rohan is rejected at 54% risk. If he cuts his other EMIs to ₹5,500 a month, or asks for ₹7.5 lakh instead of ₹9 lakh, he's approved. Here's what's happening underneath. It's a small 8 → 10 → 6 → 1 network trained on 40,000 synthetic Indian loan applications, where I deliberately gave gender zero effect on the outcome. The "why" uses Integrated Gradients against an average applicant instead of raw activations, because glowing neurons look great but explain nothing to a human. Counterfactuals come from a binary search on each input the applicant can actually change, and a fairness check re-scores every applicant with only gender flipped. That check caught the bias that accuracy missed. Retraining with gender cut out of the network dropped the effect to exactly zero, and AUC stayed at 0.92. It all runs in plain JavaScript in the browser, about 0.005 ms per decision, with no server. Code is in the comments. How does your team test lending models for bias today? Medium - https://medium.com/@anuragnagare77/neuro-lend-99cddd19faad GitHub - https://github.com/AnuragNagare/Neuro-Lend.git
0
68
Freelance Web Developer & AI Enthusiast helping brands build
Freelance Web Developer & AI Enthusiast helping brands build
Business Automation Analyst
New to Contra
Business Automation Analyst
Cover image for During my Data Analytics internship
During my Data Analytics internship at Burma Burma Restaurant & Tea Rooms, I identified that manually compiling, formatting, and auditing monthly statutory filings across 15+ outlets was consuming critical administrative hours and creating operational bottlenecks. To solve this end-to-end, So, I built an HR Statutory Compliance Engine & Real-Time Notice Portal to automate the complete workflow. ⚙️📊 The platform brings together: 🔹 Microsoft SharePoint / OneDrive Integration → Fetches centralized payroll and employee datasets through APIs with retry handling for network failures and throttling. 🔹 Dynamic Schema & Rule Mapping Engine → Automatically maps inconsistent fields such as Emp Code, Empl_Code, and Code into a standardized structure. 🔹 Automated Statutory Form Generation → Generates regional compliance registers while preserving government template layouts, rotated headers, formulas, formatting, and row structures. 🔹 Excel Engineering with Python + OpenPyXL → Handles large identifiers, prevents scientific notation, applies formulas, manages text wrapping, and dynamically prepares data rows. 🔹 Real-Time Notice & Incident Portal → A Google Apps Script-based portal for tracking statutory notices, inspections, licenses, approvals, and closure status. 📈 Business Impact ⚡ < 5 seconds — document generation per outlet 📉 85%+ — reduction in manual processing effort ✅ 0% — template formatting corruption 📊 100% — centralized compliance visibility ☁️ Cloud-synchronized and audit-ready workflows 🛠️ Technology Stack Python | Pandas | NumPy | OpenPyXL | Streamlit | Microsoft Graph API | SharePoint | OneDrive | Google Apps Script | Gmail | Excel Automation What I particularly enjoyed about this project was solving the problem at the intersection of Data Analytics + Automation + HR Operations + Regulatory Compliance. It reinforced an important lesson for me: Good automation isn't just about reducing clicks it's about building a reliable process that people can trust. This project strengthened my interest in building data-driven business automation solutions that create measurable operational impact.
0
21
Cover image for Engineered an end-to-end Invoice Processing
Engineered an end-to-end Invoice Processing Pipeline by integrating the Gemini Vision API (OCR) to dynamically extract data from Gmail PDFs. Automated vendor reconciliation, error routing, and historical audits across 15+ vendors, significantly improving accuracy and compliance
0
16
Cover image for Engineered a live mobile-responsive web
Engineered a live mobile-responsive web application using Vite, React Query, and shadcn/ui, featuring 30-second real-time data syncs and interactive visualizations (Recharts). Deployed an ensemble machine learning engine that evaluated 7 key performance variables (including H2H, player stats, and venue) to forecast live, ball-by-ball win probabilities during the WPL 2026 playoffs.
0
23
Cover image for Executive Command Centre & AI
Executive Command Centre & AI Parser Developed a full-stack app using Google Sheets as a dynamic backend, integrating Workspace APIs to automate schedule syncing and task tracking. I constructed a Gemini 3.1 Flash-Lite AI parsing engine with fuzzy-matching to automatically categorize incoming emails and messages into structured pipelines.
0
39
🧬 Merging Life & Data Science for Unique Insights
🧬 Merging Life & Data Science for Unique Insights
Cover image for Python + LLM tool that
Python + LLM tool that reads GST invoices in any layout, extracts supplier, GSTIN, line items and taxes, checks every number, and exports one Excel file with problems flagged for review.
0
36
Cover image for NLP tool that ranks candidates
NLP tool that ranks candidates against a job description using skill overlap, education match, experience and semantic similarity. Highlights Extracts skills, education and years of experience from resumes Gives each candidate a composite score using TF-IDF semantic similarity Recruiter dashboard for shortlisting
0
15
Cover image for SQL data mart for a
SQL data mart for a 1,000-patient hypertension trial, tracking enrolment, adverse events, dropout and event-free survival. Highlights Normalised schema: sites, treatment arms, patients, visits, adverse events, outcomes SQL analytics for enrolment trends, adverse-event rates by arm, and dropout rates Survival-style queries for event-free outcomes
0
20
Cover image for Executive Power BI dashboard covering
Executive Power BI dashboard covering 5 years of hotel data across cities, booking channels, managers and loyalty members. Highlights Revenue, net profit, ADR and RevPAR in one executive view Manager benchmarking and booking-channel efficiency Guest satisfaction and loyalty member analysis
0
27
GameDev, AIML Dev, CyberSec, Pixel Art, Python Automations.
GameDev, AIML Dev, CyberSec, Pixel Art, Python Automations.