DATA SCIENTIST • AI ENGINEER • BIOINFORMATICS • RESEARCHER • INNOVATOR •
DATA SCIENTIST • AI ENGINEER • BIOINFORMATICS • RESEARCHER • INNOVATOR •
I'm |
ML ENGINEER • CLOUD DATA ENGINEER • ACADEMIC PEER REVIEWER • HEALTHCARE ANALYST •
ML ENGINEER • CLOUD DATA ENGINEER • ACADEMIC PEER REVIEWER • HEALTHCARE ANALYST •
About Me
Who Am I?
A data scientist with a background in biotechnology and hands-on experience building fraud detection pipelines at a Fortune 500 healthcare company. I work across the full data stack from writing Pandas on AWS SageMaker to designing anomaly triggers and outliers that surface suspicious billing patterns at scale
My research side runs parallel with 12+ publications, 2 best paper and presentation awards across IEEE and Springer conferences, and also an independent bioinformatics work on immune checkpoint gene discovery using PCA and network analysis in R. I'm someone who moves between research and engineering without losing thread of either, and I am always more interested in what I have not figured out yet than what I already know
My Skills
What I Specialise?
Python
R
SQL
OpenAI GPT
PyTorch
TensorFlow
CI/CD
AWS
Git
Docker
Tableau
Machine Learning & AI: Anomaly Detection, Fraud Detection Pipelines, Supervised & Unsupervised Learning, Feature Engineering, Model Evaluation, Deep Learning (CNNs, Sequence Models), NLP & Transformers (Foundations)
Data Science & Analytics: Exploratory Data Analysis (EDA), Statistical Analysis, Hypothesis Testing, Data Cleaning & Preprocessing, Dimensionality Reduction (PCA), Clustering (Mclust, K-Means), KPI Tracking, Data Visualization & Reporting
Programming & Core Tools: Python (Pandas, NumPy, Scikit-learn), SQL (Athena, BigQuery, PostgreSQL, MySQL), R (ggplot2, mclust), Jupyter Notebooks, Microsoft Excel
Cloud & ML Systems: AWS (S3, Athena, Glue, Redshift Serverless, Lambda, SageMaker), ETL Pipelines, End-to-End ML Pipelines, Model Training & Deployment Workflows, Experimentation
Bioinformatics: Gene Expression Analysis, PCA-based Analysis, Mclust Clustering, Pathway Enrichment (Enrichr, KEGG), Network Analysis (STRING, Cytoscape)
Tools & Workflow: Tableau, Power BI, Looker Studio, Domo, Git, GitHub, Docker (Basics)
Experience
What I've Learned and Contributed?
Data Science Associate Analyst
The Cigna Group – Cigna Healthcare
Jan 2026 – Apr 2026
Bengaluru, Karnataka, India
Key Contributions:
- Built 12+ anomaly detection triggers across healthcare claims and provider data; designed a composite risk scoring framework using log-scaled transformations and percentile-based normalization (CUME_DIST) — improving detection coverage by 20% and contributing to an estimated $400K–$700K in potential savings
- Applied BIRCH clustering and LOF to segment providers; engineered 15+ statistical features identifying 18% high-risk providers and improving anomaly lift by 11%
- Collaborated with fraud investigators and business stakeholders to translate model outputs into prioritized investigation leads
- Engineered 40+ features from 10M+ claim records; improved model precision by 13% and deployed scoring pipelines on AWS (S3, Athena, SageMaker), flagging 17% of claims as high-risk
Undergraduate Research Assistant – Computational Bioinformatics (ML)
Bioinformatics Lab, REC · Supervised by Dr. Sujata Roy
Jan 2025 – Apr 2025
Chennai, Tamil Nadu, India
Key Contributions:
- Built an unsupervised ML pipeline on a ~49,000 feature gene expression dataset (GEO: GSE57329), applying PCA + GMM (mclust, G=9); silhouette score 0.81
- Reduced feature space by 98% (49,000 to ~1,000 genes) using sequential statistical filtering
- Identified clusters enriched for immune-related pathways; highlighted hub genes (CD4, CXCL10, FMO3) via PPI analysis and pathway enrichment (KEGG, Reactome)
AI Quantalytics – VihaanAI CyberLabs Pvt Ltd.
May 2023 – Jan 2025
Remote
Key Contributions:
- Constructed a Fully Residual CNN for brain tumor segmentation; integrated MDRNNs to enhance NLP model performance for malware classification and family prediction
- Developed and evaluated ML models for classification, clustering, and anomaly detection; built 3+ ETL pipelines
- Conducted statistical analysis across RL, Distributed ML, and NLP research tracks supporting 5+ projects
- Contributed to peer-reviewed papers for Computers in Biology and Medicine, Springer, and a patent filing
Crayon Data Pvt Ltd.
May 2023 – Aug 2023
Chennai, Tamil Nadu, India
Key Contributions:
- Contributed to NLP pipelines processing 100K+ customer records using TF-IDF/Word2Vec
- Supported BERT-based text classification achieving 85% accuracy on internal benchmarks
- Applied PCA and t-SNE for dimensionality reduction; developed sentiment and topic trend visualizations
Centre of Excellence in Biofilms, REC · Dr. Saravanan Periasamy
June 2023 – July 2023
Chennai, Tamil Nadu, India
Key Contributions:
- Synthesized EPS-based nanocomposites (AgNO₃ + SDS) from Bacillus amyloliquefaciens to inhibit oral biofilm
- Validated antimicrobial efficacy via SEM imaging and UV–Vis spectrophotometry; demonstrated 65% reduction in bacterial adhesion
- Contributed findings to a published book chapter on nanomaterial-based biofilm interventions
All Mind AI (ZRAE Global), Zion Robotics
June 2020 – Jan 2021
Chennai, Tamil Nadu, India
Key Contributions:
- Assisted in building a deep learning pipeline for forensic speech signal segmentation across 3 benchmark datasets (~2,900 audio samples)
- Contributed to experimental documentation supporting a paper submission on AI-based speech segmentation
Micro-Degree Credit Linked Degree Program — Computer Science
IIT Guwahati, Assam, India
June 2024 – Oct 2025
Guwahati, Assam
Relevant Coursework:
- MTH101 Mathematics for Computer Science
- CSE101 Introduction to Computer Science & Programming
- CSE201 Data Structures and Algorithms
- CSE202 Database Systems
- CSE301 Principles of Computer System Design & Architecture
- CSE302 Capstone Project
Bachelor of Technology — Biotechnology
Rajalakshmi Engineering College, Tamil Nadu, India
Oct 2021 – May 2025
Chennai, Tamil Nadu
Relevant Coursework:
CS / DS / Maths
- MA19153 Applied Calculus
- MA19251 Differential Equations & Vector Calculus
- MA19353 Transforms and Numerical Methods
- MA19453 Probability and Statistics
- GE19211 Problem Solving & Programming in Python
- CS19411 Python Programming for Machine Learning
Biotechnology (core)
- BT19702 Bioinformatics
- BT19201 Biochemistry
- BT19301 Microbiology
- BT19502 Molecular Biology
- BT19303 Cell Biology
- BT19602 Genetic Engineering
- BT19504 Immunology
Projects
What I've Built?
Cerebrovascular Accident Prognosis using Supervised ML
Compared Random Forest, XGBoost, and Decision Tree on clinical data to predict stroke prognosis. Applied label encoding, mean imputation, and hyperparameter tuning via grid search evaluated on AUC-ROC.
- 99.93% accuracy (Random Forest)
- 97.94% accuracy (XGBoost)
- 43,400 patient records trained & tested
Mclust-Based Gene Expression Clustering for Immune Checkpoint Analysis
Applied PCA + Mclust clustering on gene expression data to identify immune checkpoint regulators in T2D-associated atherosclerosis, validated with t-SNE and STRING network hub gene ranking.
- 49,000+ genes processed
- 9-cluster VVV model (BIC/ICL validated)
- 3 hub genes identified (Cd4, Cxcl10, Fmo3)
Healthcare Claims Fraud & Anomaly Detection Pipeline
Detected fraudulent insurance claims combining rule-based SQL flagging, provider outlier scoring with BIRCH & LOF, and anomaly ranking surfaced through a Power BI dashboard.
- 1M+ claims processed
- ~3% of claims flagged by fraud rules
- Top 1% (~128K claims) priority-queued
Log Analytics Pipeline on AWS
End-to-end serverless pipeline ingesting raw web server logs into S3, transforming via PySpark Glue ETL, and exposing 5-minute error-rate aggregations as Athena-queryable tables with real-time Power BI visualization.
- 5M+ log records processed
- ~90% detection accuracy
- 2 Glue ETL jobs (parse + aggregate)
Scientific Paper Clustering for Similarity Discovery
Grouped arXiv abstracts by semantic similarity using NLP embeddings with topic modelling per cluster and preprocessing strategy comparisons across vectorisation variants, enabling literature discovery at scale.
- 10K+ research papers
- 40% literature retrieval efficiency improved
- Multi-language preprocessing tested
Bayesian Marketing Spend Optimization (MMM)
Built a Marketing Mix Model using Bayesian inference engineering CPC/ROAS features to quantify multi-channel marketing ROI and recommend optimal budget allocation across 5 channels.
- 900 multi-channel ad records
- 5 marketing channels modelled
- Shift 10–15% budget from low-ROI channels
My Publications
What I've Contributed to the World?
Memory-Augmented Deep Recurrent Neural Networks for Long-Term Dependency Learning in Natural Language Processing
Not yet Published
An Advanced Fully Residual Convolutional Neural Network for Segmentation and Classification of Brain Tumors Across Diverse Medical Image Modalities — Computers in Biology and Medicine
Not yet Published
ROC Number: SW-16203/2023
ROC Number: SW-16851/2023
UK Design Patent, Design number: UK 6291782, Year 2023
Proceedings of Ninth International Congress on Information and Communication Technology — Springer, 2024
Perspective and Strategies on Newage Education and Creative Learning — Springer, 2024
Big Data & Edge Intelligence for Enhanced Cyber Defence — CRC Press, 2024
Big Data & Edge Intelligence for Enhanced Cyber Defence — CRC Press, 2024
Machine Learning Models and Architectures for Biomedical Signal Processing — Elsevier, 2025
Cutting-Edge Applications of Nanomaterials in Biomedical Sciences — IGI Global, 2024
Functionalisation Strategies of Silver Nanoparticles
Springer Nature, Yet to be published
My Achievements
What I've Accomplished?
Issued by IEEE 12th International Conference on Communication Systems and Network Technologies, Bhopal. Recognized for "Kernelized Deep Networks for Speech Signal Segmentation Using Clustering and AI in Neural Networks"
Issued by Kalinga University at IEEE World Conference on Communication & Computing, Raipur. Recognized for "Cerebrovascular Accident Prognosis using Supervised Machine Learning Algorithms"
Issued by the Department of Pharmaceutical Technology, UCE, BIT Campus, Anna University, Tiruchirappalli. Recognized for "A Multi-functional Aqueous Phytochemical Formulation for Minimalist Skincare and Urticaria Management"
ENIGMA 2023 — Second Prize in Poster Competition (October 2023)
Issued by Sri Manakula Vinayagar Engineering College, Puducherry. Recognized for "Optimization of the Ex-vivo Typodont Model for Dental Biofilm Associated Infections"
Participated in Symposium Vitalizz'23 at REC for Idea Presentation
Member of IEEE (Institute of Electrical and Electronics Engineers) — #98761426
Editorial Board Member for PriMera Scientific Engineering Journal (ISSN: 2834-2550)
Reviewer for Medicon Engineering Themes Journal (ISSN: 2834-7218)
Reviewed for 10+ IEEE/Springer-organized conferences indexed in Scopus — EASCT 2023, AIKIIE 2023, ICAIA 2023, ICDSIS 2024
Abstract — Brain Tumor Segmentation
Accurate and quick brain tumor diagnosis and treatment are critical for successful management. This work describes a novel technique using three deep learning architectures: GlobalNet, Multi-task Learning, and FusionNet. GlobalNet incorporates global context via a global pooling layer; Multi-task Learning allows concurrent tasks sharing knowledge; FusionNet combines information from multiple CNNs trained on various medical imaging modalities to improve accuracy and resilience of brain tumor segmentation.
Citation
S. Shreyanth, S. Niveditha and V. Kathiroli, "Accurate Brain Tumor Segmentation and Detection using Multi-Task Learning with GlobalNet and FusionNet," 2023 IEEE 12th International Conference on Communication Systems and Network Technologies (CSNT), Bhopal, India, 2023, pp. 478-485, doi: 10.1109/CSNT57126.2023.10134722.
Abstract — Speech Signal Segmentation
Speech signal segmentation is a key component of speech processing tasks including speaker recognition and emotion recognition. We propose a novel approach using kernelized deep networks with MFCC and LPC for feature extraction, spectral subtraction and Wiener filtering as preprocessing, PCA and LDA for dimensionality reduction, and spectral clustering using the graph Laplacian. Results on TIMIT and BURN corpus demonstrate superior segmentation accuracy with reduced computational complexity.
Citation
S. Niveditha, S. Shreyanth, V. Kathiroli, P. Agarwal and S. Ram Abishek, "Kernelized Deep Networks for Speech Signal Segmentation Using Clustering and Artificial Intelligence in Neural Networks," 2023 IEEE 12th International Conference on Communication Systems and Network Technologies (CSNT), Bhopal, India, 2023, pp. 667-674, doi: 10.1109/CSNT57126.2023.10134609.
Abstract — Cluster-Based Grid Computing
Grid computing based on clusters improves efficacy of wireless network data transmission. This study examines cluster-based grid computing, routing analysis protocols, and deep learning to optimize data transmission. The proposed method utilizes clusters to distribute computing duties, a novel routing analysis protocol that dynamically adapts to network conditions, and deep learning for intelligent data routing. Results show substantial enhancements in throughput, latency, and energy consumption.
Citation
S., Shreyanth & S., Niveditha. (2023). CLUSTER-BASED GRID COMPUTING ON WIRELESS NETWORK DATA TRANSMISSION WITH ROUTING ANALYSIS PROTOCOL AND DEEP LEARNING. International Journal of Advanced Research. 11. 517-534. 10.21474/IJAR01/17096.
Abstract — Neural Image Caption Generation
We propose a novel model for neural image caption generation with visual attention to help visually impaired individuals access online images. Our model uses CNNs and RNNs to convert image content into aural descriptions. The model operates in two phases: a text-to-speech API converts image content into textual description, then the extracted description is converted to audio. This research promotes digital inclusion and equality for millions of visually impaired people.
Citation
Priyanka Agarwal, Niveditha S, Shreyanth S, Sarveshwaran R, Rajesh P K, "Neural Image Caption Generation with Visual Attention: Enabling Image Accessibility for the Visually Impaired," International Journal of Scientific Research in Science, Engineering and Technology (IJSRSET), Print ISSN: 2395-1990, Online ISSN: 2394-4099, Volume 10, Issue 3, pp.562-575, May-June-2023.
Abstract — Machine Learning in VLSI Design
Machine learning has made an impact in the microchip area and will eventually supplant current VLSI design concepts. Design creation has been automated by substituting time-consuming traditional concepts developed by experts. Several major firms including IBM and Intel have their own CAD division for design automation. This article discusses the importance of ML application in VLSI chip design and development and how ML-oriented BIST was implemented.
Citation
Shreyanth, S., Harshitha, D.S. & Niveditha, S. Implementation of Machine Learning in VLSI Integrated Circuit Design. SN COMPUT. SCI. 4, 137 (2023).
Abstract — Fintech, Crisis, and Marketing
Fintech has revolutionized how consumers access and manage finances. In times of crisis, retaining customers is crucial for financial firms. This research explores how technology-driven financial firms adapt their marketing strategies to retain customers during crises, examining methods using CRM systems, customer data analysis, Pareto principle, Kano model, RFM analysis and Entropy-weighted CLV for identifying and prioritizing valuable customers.
Citation
Shreyanth, S., Suwetha, P. T., Kathiroli, V., Niveditha, S., & Jayaprakash, H. (2023). Fintech, Crisis, and Marketing: How Technology-Driven Financial Firms Adapt Their Approach to Retain Customers. Proceedings of the International Conference on Emerging Trends in Business & Management (ICETBM 2023), 309–321. Atlantis Press. https://doi.org/10.2991/978-94-6463-162-3_28
Abstract — Cerebrovascular Accident Prognosis
This study compares the predictive performance of three supervised learning algorithms — Random Forest, Decision Tree, and XGBoost — for cerebrovascular disease prediction. Random Forest achieved the highest prediction accuracy of 99.93%, XGBoost showed 97.94%, and Decision Tree obtained 97.53%. These results outperform state-of-the-art methods and demonstrate the potential of supervised learning for accurate cerebrovascular prediction in clinical practice.
Citation
B. Sethuraman and S. Niveditha, "Cerebrovascular Accident Prognosis using Supervised Machine Learning Algorithms," 2023 World Conference on Communication & Computing (WCONF), RAIPUR, India, 2023, pp. 1-8, doi: 10.1109/WCONF58270.2023.10235122.
Abstract — Malware Classification
This study presents a unique method for predicting malware category and family using machine learning, leveraging the Cuckoo environment and automated feature selection via the Boruta method. The KNN technique was combined with dynamic analysis of malware samples within the Cuckoo environment. Test Case 3 (Automated Feature Selection + Cross-Validation) showed outstanding specificity (90%), precision (93%), recall (96%), and F1-Score (92%) for phishing detection.
Citation
Niveditha S & Rr, Prianka & K, Sathya & S, Shreyanth & Subramani, Nandhagopal & Deivasigamani, Balakrishnan & S, Karthikeyan. (2024). Predicting Malware Classification and Family using Machine Learning: A Cuckoo Environment Approach with Automated Feature Selection. Procedia Computer Science. 235. 2434-2451. 10.1016/j.procs.2024.04.230.
Abstract — NLP Resume Screening
This study presents an AI-driven resume screening system based on NLP using a proprietary spaCy NER library to extract key data from resumes including names, organizations, job titles, skills, and education. The system employs data preprocessing techniques including tokenization, NER-based scoring, and ranked candidate listing. Results showed the NLP-driven solution surpassed traditional approaches in accuracy, efficiency, and fairness, reducing time and effort required for resume screening.
Citation
Sarveshwaran R., Karthikeyan S., Cruz M.V., Shreyanth S., Niveditha S., Rajesh P.K. (2024). NLP-Based AI-Driven Resume Screening Solution for Efficient Candidate Selection. In: Yang, XS. et al. (eds) Proceedings of Ninth International Congress on Information and Communication Technology. ICICT 2024. Lecture Notes in Networks and Systems, vol 1012. Springer, Singapore. https://doi.org/10.1007/978-981-97-3556-3_29