Building Explainable
AI Systems
for High-Stakes
Decisions.
Machine Learning Engineer focused on designing
scalable, interpretable, and production-ready AI
systems across fintech and geospatial domains.
.png)
THE MIND BEHIND
INTELLIGENT SYSTEMS
I design machine learning systems that don’t just predict — they explain. In high-stakes domains like fintech and geospatial intelligence, black-box models fail where trust is critical.
My work focuses on building transparent, scalable, and production-ready AI systems that bridge the gap between model performance and real-world decision-making. From large-scale data pipelines to explainable decision systems, I aim to create AI that stakeholders can understand, trust, and act upon.
Explainability Meets
High-Stakes Intelligence
Highlighting key systems that combine predictive power with full model interpretability.
Explainable Credit Risk Platform
End-to-end ML pipeline on 1.3M+ financial records with SHAP-based explainability for regulatory transparency.
T2-Hydro: Geospatial Intelligence
Deep learning system for drought forecasting using a CNN-LSTM hybrid architecture with 85% predictive accuracy.
Deep Expertise in
AI & Data Engineering
I specialize in building production-grade AI systems that are transparent, scalable, and built for real-world impact.
Machine Learning Systems
Designing end-to-end ML pipelines from data ingestion to deployment, optimized for performance and scalability.
Explainable AI (XAI)
Implementing SHAP and interpretability frameworks to ensure transparency and regulatory alignment.
Data Engineering & Pipelines
Building robust ETL systems for processing large-scale datasets.
Cloud & Deployment
Deploying production-ready applications using AWS, Docker, and Streamlit.
Advancing the Field Through
Research, Patents, and Publications
My work extends beyond system implementation into the realm of academic research and innovation. I focus on making AI more accessible and transparent through publications and patented systems.
A Multimodal Narration System for Enhancing Accessibility for the Visually Impaired
Published in IC-CGU 2025: Developed an AI-driven OCR/TTS system for real-time visual-to-audio narration.
- Achieved 90.96% OCR accuracy and 92.77% F1-score; optimized TTS module for 4.14/5 mean user rating.
A Distributed Meta-Learning Framework for Rapid Anomaly Adaptation in Non-IID Financial Networks
Developed a novel Federated Meta-Learning framework that replaces global model convergence with meta-initialization, enabling few-shot anomaly detection (5 samples) in highly non-IID financial environments.
- Introduced a meta-gradient exchange mechanism with bi-level optimization, achieving faster edge adaptation and significantly reducing communication overhead and latency in decentralized systems
EXPLAINABILITY IS THE
NEW ELEGANCE
Where data meets intelligence, and decisions become transparent. I believe the future of AI lies not just in performance, but in trust — systems that are interpretable, reliable, and built for real-world impact.
How I Build
Intelligent Systems
A systematic approach to building AI that stakeholders can understand, trust, and act upon.
Data Understanding
Deep exploration, validation, and preprocessing of complex datasets.
Model Design
Selecting and engineering models optimized for real-world constraints.
Explainability Layer
Integrating SHAP and interpretability frameworks for transparency.
Deployment
Building scalable, real-time systems using cloud and containerization.
Exposure & Experience
What I do beyond building systems — leadership, exposure, and real-world experience.
Geospatial Intelligence
Applying deep learning to environmental risk management and satellite data analysis.
Fintech Innovation
Building explainable credit risk models and transparent AI systems for financial institutions.
Research & Publication
Active contributor to IEEE conferences with focus on accessible AI and meta-learning.
Geospatial Intelligence
Applying deep learning to environmental risk management and satellite data analysis.
Fintech Innovation
Building explainable credit risk models and transparent AI systems for financial institutions.
Research & Publication
Active contributor to IEEE conferences with focus on accessible AI and meta-learning.
Let’s Build Intelligent Systems Together
Open to internships, collaborations, and impactful AI projects.