Nitin Datta
Movva

Building intelligent solutions with AI, ML, and modern web technologies.
Redefining limits, pushing boundaries, creating impact.
I'm a Machine Learning Engineer and Software Developer with a strong background in building practical, high-impact AI systems. With experience spanning hybrid retrieval pipelines, LLM fine-tuning, and classical ML modeling, I've improved model accuracy, system reliability, and end-to-end automation across industries including healthcare, semiconductors, and enterprise software. My work focuses on turning complex data and ML techniques into solutions that deliver clear business value.
On the engineering side, I develop modern, scalable applications using technologies like Next.js, React, FastAPI, and cloud-native tooling. I've led major migrations, built CI/CD workflows, and created orchestration layers that simplify how teams interact with ML systems. I enjoy crafting products that pair clean software engineering with thoughtful machine learning design—tools that feel simple to use but are backed by solid technical foundations.
End-to-end ensemble trading framework that trains, backtests, and promotes validated strategies to live execution. Features a ranked ensemble learning system with dynamic strategy evaluation, supporting mean reversion, momentum, arbitrage, and AI-driven trading strategies with MongoDB data storage and comprehensive backtesting capabilities.
10th place solution to BirdCLEF - Birdcall Identification Competition. Ensemble of CNN-based models with custom augmentations, probability calibration, and geofencing post-processing for bird sound detection in long audio recordings.
End-to-end data engineering pipeline for real-time data ingestion and processing. Orchestrates data flow from API sources through Apache Airflow, streams via Apache Kafka, processes with Apache Spark, and stores in PostgreSQL and Cassandra databases, all containerized with Docker.
I'm always open to discussing new projects, creative ideas, or opportunities to be part of your visions.
When I'm not building applications or training models, I enjoy exploring the world through travel and outdoor adventures. I'm also interested in cooking, which allows me to experiment with flavors and create something new. These experiences fuel my creativity and provide fresh perspectives that I bring back to my work.
I believe that the best solutions come from a well-rounded perspective, and my hobbies help me stay curious, creative, and connected to the world around me.

"The magic you are looking for is in the work you are avoiding"
— Chris Williamson
Financial Markets
Burn 500+ calories daily
Tracking via Apple Watch through hikes, inclined walks, and stairmaster
Contribute to open source
Regular contributions to community projects
1 hour daily on dream project
Dedicated time for building an algo trading system
30%+ returns on portfolio
Financial growth through strategic investments

Deep Work
Cal Newport

Thinking, Fast and Slow
Daniel Kahneman

Advances in Financial Machine Learning
Marcos López de Prado
I'm always open to discussing new projects, creative ideas, or opportunities to be part of your visions.