logo True scholar network
Back to Top

About

Rahul Arulkumaran is a dynamic AI Engineering Manager at Yuma, with an extensive background in artificial intelligence, machine learning, data science, and blockchain technologies. With a research-driven mindset, he has secured two patents and authored three research papers, demonstrating his ability to innovate and contribute meaningfully to the tech industry. His expertise encompasses the entire data science pipeline, including data engineering, exploratory data analysis, predictive modeling, natural language processing (NLP), time series forecasting, and AI-driven decision-making. He has a deep understanding of cloud computing, working extensively with AWS, Snowflake, Terraform, and Docker to build scalable and efficient data pipelines. Rahul holds a Master of Science in Data Science from the University at Buffalo, where he honed his skills in advanced analytics and blockchain technology, working as a research assistant at Blockchain ThinkLab under Professor Bina Ramamurthy. Before joining Yuma, Rahul was an AI/ML Engineer Fellow at Foundry, where he played a pivotal role in developing machine learning models to predict S&P 500 prices using BitTensor. His work led to the establishment of a new business line projected to generate millions in annual revenue. He successfully architected and optimized a subnet with 100% utilization of 256 miner slots, demonstrating his expertise in distributed computing and blockchain-based AI applications. His tenure at Foundry also included roles as Data Engineer III and Data Engineer II, where he built highly scalable ETL pipelines handling over 10TB of historical data and 10GB of daily data, integrating multiple AWS services such as S3, Glue, Athena, ECS, and Kinesis. He also played a key role in implementing Mulesoft for ETL and API development, significantly improving internal data access efficiency. His contributions reduced API response times by 95%, streamlined data reconciliation processes, and enhanced data availability across teams. Rahul's entrepreneurial mindset is evident in his experience as the CEO and Co-Founder of Widhya, an ed-tech startup that grew from 650 to 15,000 users in just three months, leveraging data-driven growth strategies. His leadership in the Web3 space includes co-founding NFT Garage and Stonk Society, a successful NFT project that raised $200,000 in sales. Additionally, he worked as a part-time developer for SpiritSwap, designing robust APIs and portfolio analytics tools for blockchain users. His ability to bridge AI, blockchain, and data science has allowed him to contribute meaningfully to various decentralized finance (DeFi) projects across Ethereum, Fantom, Avalanche, and Polygon ecosystems. Beyond his professional roles, Rahul actively contributes to the technology community. He is a member of the Forbes Technology Council and an early adopter at the AI Frontier Network, highlighting his influence in AI and blockchain discussions. As a technical reviewer for O'Reilly, he ensures the quality and accuracy of AI and data science publications. He is also a member of prestigious organizations like the International Association of Engineers (IAENG) and the Association of Data Scientists, reflecting his commitment to continuous learning and thought leadership. Rahul’s technical expertise extends to impactful projects such as the Bitcoin Price Predictor, which utilized linear regression and data visualization techniques to forecast cryptocurrency prices, securing him a top-five position in a hackathon. Another significant project, the Depression Detection Bot, leveraged NLP and AI to assess mental health risks and suggest appropriate treatments. His ability to apply AI solutions to diverse fields—from finance to mental health—demonstrates his versatility and problem-solving capabilities. With an impressive skill set spanning machine learning, AI, blockchain, Web3 development, cloud computing, and data engineering, Rahul continues to push the boundaries of technology. His leadership, innovative mindset, and technical prowess make him a significant contributor to the evolving landscape of artificial intelligence and blockchain applications.

View More >>

Skills

Experience

placeholder
AI Engineering Manager

Yuma AI

Jan-2025 to Present

Education

placeholder
University at Buffalo, New York

M.SC in Data Science

Passout Year: 2022

Publication

  • dott image November, 2021

Analyzing Information Asymmetry in Financial Markets Using Machine Learning

Journal : IJPREMS - International Journal of Progressive Research in Engineering Management and Science

When it comes to financial markets, information asymmetry, which occurs when different players have access to differing degrees of knowledge, may result in inefficient markets, skewed prici...

Peer-Reviewed Articles

ASD-Pipeline: An Ensemble Machine Learning Framework Integrating Feature Selection, Behavioural Clustering, and Class Rebalancing for Accurate Autism ...

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition characterized by a variety of behavioral and cognitive patterns. Early and precise detection is critical in enabling timely interventions. Conventional classification models frequently exhibit poor generalization due to irrelevant features, unstructured behavioral data, and severe class imbalance. Despite current advances in machine learning for ASD detection, current models do not integrate adaptive feature selection, behavioral grouping, or imbalanced class handling in a unified, end-to-end pipeline. The lack of incorporation frequently results in suboptimal performance and limited interpretability. This study proposes a new ensemble-based framework called ASD-Pipeline, which integrates flexible feature selection, hybrid clustering, synthetic minority oversampling, and ensemble voting classification to improve the predictive performance for ASD identification. The proposed ASD-Pipeline framework uses a five-stage process to improve the accuracy of autism spectrum disorder prediction. First, the dataset is normalized utilizing Min-Max scaling to guarantee that the feature ranges remain consistent. Next, feature selection is performed utilizing FlexiFeat, an ensemble method integrating filter-based (CfsSubsetEval with BestFirst), wrapper-based (WrapperSubsetEval with GreedyStepwise), and embedded (ReliefF with Ranker) techniques to maintain only the most pertinent feature. The ClusterGroup stage uses K-Means clustering (k=5) and DBSCAN improvement (ε=0.5, minPts=3) within each cluster to create behavioral groups and remove outliers. The ReBalance stage uses Cluster-SMOTE to tackle class imbalance by producing synthetic samples for the minority class and a balanced dataset. Finally, the ASDClassifier stage involves training an ensemble of Logistic Regression, Support Vector Machine, and Gradient Boosting classifiers that are combined using soft voting. Metrics used to assess the model include accuracy, precision, recall, F1-score, and Matthews Correlation Coefficient (MCC). The proposed ASD-Pipeline surpassed existing models, achieving a significantly higher accuracy of 96.18% compared to previous techniques ranging from 76.80% to 90.60%. It also scored 91.51% precision, 91.63% recall, 95.57% F1-score, and 92.51% specificity. These findings emphasize the pipeline's efficacy in enhancing generalization and tackling difficulties such as feature relevance, behavioral grouping, and class imbalance for ASD prediction. The ASD-Pipeline offers a reliable, interpretable, and modular machine learning solution for ASD prediction. Its incorporated method tackles critical challenges in feature relevance, behavioral variability, and data imbalance, rendering it a promising tool for healthcare practitioners and researchers seeking data-driven insights into early ASD detection.

Projects

Mar-2018 to Dec-2018

Bitcoin Price Predictor

This was a project that I worked on during the Hackathon organised by JNTUHCEH. It's a Bitcoin Price Predictor that uses visualisation techniques to figure out the important fields in a Dataset to perform a Data Science. Finally, I used the Linear Regression algorithm to predict prices of Bitcoins on any given day. This project helped me secure the 5th position in the Hackathon
...see more

Certificates

Issued : Mar 2021
  • dott image By : University of P...
  • dott image Event : Coursera Inc.
Fundamentals of Quantitative Modeling
https://coursera.org/verify/LQDX6JZV9R2G

Membership

dott image
Member

International Association of Engineers (IAENG)

From year 2023 to Present

Scholar9 Profile ID

S9-022025-2409766

Publication
Publication

(1)

Review Request
Article Reviewed

(21)

Citations
Citations

(10)

Network
Network

(3)

Conferences
Conferences/Seminar

(0)

Academic Identity

Scopus Profile
Scopus Profile
Google Scholar Profile
Google Scholar Profile
Academia Profile
Academia Profile
Research Gate Profile
Research Gate Profile
YouTube Link
YouTube Link

Authors

Shreyas Mahimkar
Shreyas Mahimkar

Data Scientist at Li...
Sumit Shekhar
Sumit Shekhar

JPMC Product Leaders...
Rahul Arulkumaran
Rahul Arulkumaran

AI Engineering Manag...