Hello, I'm

Harshith Narasimhamurthy

|

Transforming data into actionable insights through geospatial intelligence & ML

View My Work Get In Touch
0 Years Experience
0 Projects Delivered
0 Accuracy Rate
Scroll Down
Get to know me

About Me

I'm a AI enabled Data Scientist with a passion for transforming complex data into strategic business solutions. With 3 years of hands-on experience in building and deploying production-grade ML models, I specialize in geospatial intelligence, time-series forecasting, and cloud-native analytics.

My journey began with a Gold Medal in M.Sc. Agricultural Statistics and has evolved into driving measurable impact across finance, agriculture, and enterprise analytics. I thrive on solving challenging problems from predicting crop health using satellite imagery to optimizing bank branch locations through spatial analytics.

Currently at SatSure Analytics, I own end-to-end ML solutions that support banking, insurance, and agribusiness, translating geospatial and temporal data into high-impact insights. My work has delivered double-digit improvements in operational efficiency and helped businesses make data-backed decisions with confidence.

Customer Champion

SatSure Analytics

Gold Medalist

M.Sc. Statistics

AIR-6 Rank

B.Sc. Agriculture

Harshith Narasimhamurthy
My Journey

Work Experience

July 2023 - Present

Data Analyst

SatSure Analytics • Bengaluru, India

  • Owned and delivered production ML solutions supporting banking, insurance, and agribusiness use cases, translating complex geospatial and temporal data into high-impact business insights
  • Partnered with cross-functional teams to operationalize predictive and analytical models, enabling data-backed decision-making for expansion planning, risk assessment, and customer targeting
  • Applied statistical experimentation (A/B testing, hypothesis testing) and advanced SQL analytics to drive double-digit improvements in sales and operational efficiency
  • Designed and automated scalable ETL/ELT pipelines using Apache Airflow and Apache Iceberg on AWS, improving data reliability and reducing processing latency by ~25%
  • Recognized as Customer Champion for consistently delivering high-quality, impact-driven analytics solutions
Jan 2023 - Mar 2023

Machine Learning Researcher

Western Sydney University • Sydney, Australia

  • Researched and implemented time-series forecasting models (LSTM, ARIMA) for crop yield prediction
  • Designed preprocessing and validation workflows, improving forecasting accuracy by ~12% while reducing manual analytical effort
  • Published research papers on deep learning applications for agricultural price forecasting
Featured Work

Data Science Projects

Bank Branch Expansion

Location Intelligence & Geospatial Analytics

Watch Demo
Machine Learning K-Means Geospatial

Bank Branch Expansion using Location Intelligence

Identified high-potential, profitable locations for new bank branches through advanced geospatial analytics and ML modeling.

82% Accuracy
35% Cost Reduction
  • Deep EDA and feature engineering on large-scale demographic datasets
  • K-Means clustering for market segmentation
  • Random Forest & Linear Regression for profitability prediction
Watch Demo

Crop Health Monitoring

Remote Sensing & Time Series Analysis

Watch Demo
Time Series NDVI K-Means

Real-Time Crop Health Monitoring

Built a high-resolution crop health intelligence system using time series NDVI for risk assessment and insurance validation.

95% Accuracy
10m Resolution
  • Statistical time-series analysis on multitemporal NDVI data
  • K-Means clustering to classify crop health (Good/Average/Poor)
  • Adopted by banks and insurers for loan underwriting and yield assurance
Watch Demo

Cross-Industry BI Solutions

Advanced SQL Analytics

SQL Business Intelligence Analytics

Cross Industry Business Intelligence Solutions

Designed a reusable, SQL-driven analytics repository solving real-world problems across Banking, Retail, Logistics, and Healthcare.

  • Advanced SQL using Window Functions, CTEs, recursive queries
  • Domain-specific KPIs: RFM, basket analysis, customer segmentation
  • Scalable framework adaptable to diverse business environments

F1 AI SQL Assistant

Natural Language to SQL · LLM · Streamlit

Live App
LLM Text-to-SQL Groq · LLaMA 3.3 Streamlit

F1 AI SQL Assistant — Text to SQL

A production-grade natural language interface that lets anyone query Formula 1 race data in plain English — no SQL required.

  • Applied prompt engineering with Groq's LLaMA 3.3 70B to auto-generate, execute, and self-correct SQL queries — demonstrating end-to-end LLM integration with real database backends
  • Implemented full conversation memory, smart clarification logic, and SQL safety guardrails (SELECT-only enforcement), ensuring production-ready reliability
  • Delivered a zero-barrier analytics experience — translating complex F1 race data into precise human-readable answers, proving how LLMs can democratize data access
Live App

RNN Sentiment Analysis

Deep Learning · NLP · Streamlit

Live App
RNN / LSTM NLP Deep Learning Streamlit

Sentiment Analysis on Movie Reviews

A deep learning NLP application that classifies movie review sentiment, helping viewers make informed viewing decisions based on collective audience opinion.

  • Built and trained a Recurrent Neural Network (RNN/LSTM) on the IMDB dataset, applying tokenization, sequence padding, and embedding layers to capture contextual sentiment patterns in text
  • Deployed as a live Streamlit app, showcasing the full ML lifecycle — from model training to real-time inference on user-submitted reviews
  • Demonstrated core NLP skills — text preprocessing, sequential modelling, and binary classification — with practical application in consumer decision-making
Live App

AI News Automation

n8n · Groq LLM · Telegram · RSS

n8n LLM Automation Groq Telegram

Automated Morning Intelligence Newsletter

An AI-powered daily workflow that aggregates multi-category news, intelligently structures it using a Groq LLM, and delivers a curated Telegram newsletter — every morning before market open.

  • Engineered an end-to-end n8n automation pipeline ingesting RSS feeds across Finance, AI, Politics, Sports, and World events — with a Groq LLM categorizing and structuring headlines into a clean, readable digest
  • Reduced daily information-gathering time significantly by replacing manual multi-platform browsing with a single, AI-curated briefing — demonstrating practical productivity impact of AI micro-automations
  • Proved that thoughtful, well-scoped LLM integrations deliver consistent real-world value — an important principle in responsible, production-minded AI development
Technical Expertise

Skills & Technologies

Core Competencies

Python & SQL 95%
Machine Learning (Scikit-learn, Random Forest) 90%
Geospatial Analytics & NDVI 92%
AWS (S3, Athena, EC2) 85%
Apache Airflow & Iceberg 80%
Data Visualization (Tableau, Matplotlib) 88%

Tools & Technologies

Python
SQL
AWS
Scikit-learn
Pandas
Tableau
Apache Airflow
Git/GitHub
PostgreSQL
LSTM/ARIMA
NumPy
Seaborn

Certifications & Publications

Generative AI with Langchain & Huggingface

Advanced Python for Data Science

HackerRank SQL Certification

Published: Neural Networks for Price Forecasting

Academic Background

Education

M.Sc. Agricultural Statistics

Anand Agricultural University

2021 - 2023 • Anand, Gujarat

Gold Medalist

B.Sc. Agriculture

University of Agricultural Sciences

2017 - 2021 • Bangalore

All India Rank - 6
Get In Touch

Let's Connect

I'm always open to discussing new projects, creative ideas, or opportunities to be part of your vision. Feel free to reach out!

Email

harshithnchandan@gmail.com

Phone

+91 9663918804

Location

Bangalore, India