● Applied Geospatial AI & Environmental Intelligence

Reading the planet in patterns.

I’m Aniruddha, a computer science researcher at PDPM IIITDM Jabalpur exploring how satellite remote sensing, spatio-temporal deep learning, and ground sensor fusion can transform planetary observation into clearer environmental decisions.

Earth Texture: NASA Blue Marble / SVS ↗
9.33

PhD Coursework CPI

PDPM IIITDM Jabalpur (Doctoral Research in CSE)

23 Mo.

Assistant Professor

University-accredited higher education faculty tenure

State Honor

Medhavi Vidyarthi Award

Conferred by Chief Minister for top academic distinction

Multi-Band

Satellite Sensor Fusion

Copernicus Sentinel-5P, MODIS AOD & IoT Ground Stations

Test the models in real time.

Explore live mathematical simulations, satellite multi-spectral band decompositions, and deep spatio-temporal neural architectures.

Meteorological & Emission Inputs

Adjust parameters to observe non-linear atmospheric dispersion dynamics

PRESET SCENARIOS
Ambient Temperature: 12 °C
Relative Humidity: 82 %
Boundary Layer Wind: 2 km/h
Urban Activity & Traffic: 85 %
285
PREDICTED AQI Severe / Inversion
PEAK PM2.5 154 µg/m³
MODEL R² CONFIDENCE 95.8% R²
WHO Safe Threshold (15 µg/m³)
EXPLAINABLE AI: REAL-TIME SHAP FACTOR ATTRIBUTION Physics-Guided Feature Weights
Thermal Inversion
Urban Activity
Humidity Particle Growth
Wind Dispersion
SPLIT: TRUE COLOR (LEFT) vs SPECTRAL INDEX (RIGHT)

NDVI (Normalized Difference Veg. Index)

Biophysical remote-sensing index measuring the difference between near-infrared (which vegetation strongly reflects) and red light (which vegetation absorbs).

NDVI = (B8_NIR - B4_Red) / (B8_NIR + B4_Red)
Satellite Sensor: Sentinel-2 MSI
Spatial Resolution: 10m Ground Sampling
Revisit Frequency: 5 Days Constellation
Spectral Range: 443 nm – 2190 nm

Photometric Diagnostic

Drag the interactive split handle to compare visible human RGB spectrum against deep multispectral infrared penetration.

STAGE 01

Multi-Modal Ingestion

Sentinel-5P TROPOMI & MODIS AOD paired with CPCB ground IoT monitors.

Tensor: [B, 72, 48, 7]
STAGE 02

Spatio-Temporal Graph

Dynamic directed graph weighted by distance & boundary-layer wind advection.

Graph: G(V, E, W_t)
STAGE 03

ST-GNN & Attention

Spatial convolution & multi-head temporal attention with physics losses.

Attention: 8 Heads
STAGE 04

Uncertainty & Advisory

Calibrated Bayesian uncertainty intervals for municipal clean city actions.

Output: ŷ ∈ ℝ^{N × 24}

01 / Multi-Modal Data Ingestion

CLICK ANY STAGE TO INSPECT

Synchronizes continuous satellite spectral granules (Sentinel-5P Level 2 tropospheric NO₂, MODIS AOD) with hourly ground IoT monitoring stations from the Central Pollution Control Board (CPCB) and ERA5 meteorological boundary-layer reanalysis.

Input Tensor: X ∈ ℝ^{B × T × N × F}
B: Batch Size (32)
T: Temporal Window (72 Hours)
N: Ground Monitoring Nodes (48)
F: Features (PM2.5, NO₂, Temp, RH, Wind_U, Wind_V, AOD)

Scientific papers & preprints.

Investigating machine learning architectures for environmental time-series, remote sensing, and computing education.

Deep Spatio-Temporal Graph Neural Networks for Satellite-Calibrated Urban Air Quality Forecasting

Aniruddha, Research Collaborators & Mentors — Doctoral Research, IIITDM Jabalpur
Working Paper 2025

Presents an end-to-end spatio-temporal framework coupling Sentinel-5P TROPOMI satellite imagery with sparse ground IoT air sensors across central Indian urban corridors. Formulates dynamic adjacency matrices driven by boundary-layer wind advection, demonstrating a 23.4% reduction in root mean square error during extreme winter thermal inversion events.

Cross-Scale Remote Sensing and Ground-Sensor Fusion for Surface Particulate Matter Estimation

Aniruddha — Environmental Informatics & Geospatial Preprints
Preprint 2024

Investigates multi-sensor fusion combining high-revisit Moderate Resolution Imaging Spectroradiometer (MODIS) Aerosol Optical Depth (AOD) with micro-scale meteorological variables to resolve continuous ground-level PM2.5 concentrations over complex topographical terrain.

An Empirical Study of Machine Learning Architectures in High-Dimensional Urban Environmental Time Series

Aniruddha — Master of Technology Research Thesis, RGPV Bhopal (8.89 CGPA Distinction)
MTech Thesis 2024

Benchmarking classic ensemble techniques against deep recurrent and attention models for multi-horizon urban air quality prediction. Identified seasonal lag dependencies and validated feature attribution models to enhance public environmental transparency.

Pedagogical Systems in Higher Technical Education: Bridging Theory and Practical Implementation

Aniruddha — Perspectives in CS Pedagogy (From 23 Months Assistant Professorship)
Faculty Study 2023

Synthesizes two years of academic classroom leadership mentoring university undergraduates across Database Management Systems, Data Mining, and Operating Systems. Evaluates interactive software scaffolds to accelerate student mastery of abstract algorithmic structures.

Portrait of Aniruddha
ANIRUDDHA / INDIA 23.17° N, 80.02° E

A systems-first methodology

From raw planetary signals to actionable science.

My research sits at the intersection of computational learning systems and earth-scale environmental dynamics. As a doctoral scholar at PDPM IIITDM Jabalpur, I focus on the challenge of spatial and temporal resolution gaps: how to harmonize coarse, planet-wide satellite observations with hyper-local, continuous ground sensors to forecast pollution, vegetation changes, and environmental stress.

Having spent nearly two years in university lecture halls as an Assistant Professor, I believe deeply that rigorous research must also be communicated clearly, transparently, and with public purpose.

PhD Computer Science
PDPM IIITDM Jabalpur (CPI: 9.33)
23 mo. University Faculty
Assistant Professor
State Merit Medhavi Vidyarthi Award
by Chief Minister

Rigorous foundation & faculty experience

A lifelong commitment to scholarship.

Academic Qualifications

01

2025 — Present

Doctor of Philosophy (CSE)

PDPM Indian Institute of Information Technology, Design and Manufacturing, Jabalpur

Doctoral Coursework CPI: 9.33 / 10.0
PhD

2024

Master of Technology (CSE)

Rajiv Gandhi Proudyogiki Vishwavidyalaya, Bhopal

8.89 CGPA (Distinction)
MTech

2020

Bachelor of Engineering (CSE)

Rajiv Gandhi Proudyogiki Vishwavidyalaya, Bhopal

8.48 CGPA (First Class with Honours)
BE

2016 & 2014

Higher Secondary & Secondary Education

Madhya Pradesh Board of Secondary Education, Bhopal

Class 12: 90.00% · Class 10: 85.67%
Merit

Faculty & Pedagogy Tenure

02

JUL 2023 — JUN 2025 (23 MONTHS)

Assistant Professor

Unique College Parasia
Affiliated to Raja Shankar Shah University, Chhindwara

Delivered undergraduate curricula in Operating Systems, Database Management Systems, Data Mining, and Artificial Intelligence. Mentored hundreds of students in computer science foundations and project engineering.

Database Systems Data Mining Operating Systems AI / ML Pedagogy Academic Mentorship
State Merit Award

Medhavi Vidyarthi Award

Conferred by Shivraj Singh Chouhan (former Chief Minister of Madhya Pradesh) in recognition of top academic distinction.

Industrial Engineering

Indian Railways Telecom Training

30-Day intensive vocational training in Control Communication, RailNet, and Exchange Systems in the Jabalpur Division, West Central Railway.

Continuous Excellence

AI/ML Capacity Building & TCS iON

Completed two-week national capacity building in Artificial Intelligence & Machine Learning, plus TCS iON Career Edge Young Professional credential.

Applied intelligence with a public purpose.

Bridging predictive algorithms, structured databases, and robust software architectures.

02 / EDUCATION SYSTEMS 2020

Educational platform

E-learning desktop system.

A Windows-based learning architecture built with Visual Studio and MS SQL Server that consolidates programming and data-structure curricula into one structured environment.

Java Visual Studio MS SQL Server Pedagogy
03 / MEDIA SYSTEMS 2020

Desktop media engine

AS Media Player.

A Java-based media playback engine designed around low-latency audio/video stream buffering and codec integration using the Java Media Framework (JMF API).

Java JMF API GUI Architecture

A practical base for scientific exploration.

Mathematical tools, algorithmic frameworks, and collaborative academic habits.

01

Deep Learning & AI

Graph Neural Networks, Temporal Attention, Time-Series Forecasting, and Explainable AI (SHAP).

PYTHON · PYTORCH
02

Geospatial & Remote Sensing

Satellite multispectral band processing (Sentinel-5P, MODIS AOD, Sentinel-2), spatial advection modeling.

REMOTE SENSING · SATELLITE ML
03

Data Engineering & Systems

Relational databases, high-throughput ingestion pipelines, schema optimization, and SQL analytics.

MYSQL · MS SQL SERVER
04

Teaching, Mentorship & Pedagogy

University-level instructional design, technical curriculum engineering, and academic leadership.

HIGHER EDUCATION · MENTORSHIP

The Research Philosophy

Observe. Model. Act.

My doctoral journey at IIITDM Jabalpur is centered on building learning systems that make planetary change intelligible, reliable, and actionable for human communities.

01 Observe

Ingest multi-spectral satellite granules and ground IoT telemetry.

02 Model

Construct physics-guided spatio-temporal neural architectures.

03 Act

Deliver calibrated forecasts for municipal and clean-air policy.

Let’s collaborate on questions with real-world scale

Have a question worth mapping?

aniruddha364@gmail.com

Based in Madhya Pradesh, India (PDPM IIITDM Jabalpur / Chhindwara)
Phone: +91 70001 99835  ·  +91 97522 44073
Available for doctoral research collaborations, academic exchanges, and geospatial AI partnerships.

Download Full Curriculum Vitae (PDF)