SUMMARY:

Around 8+ years of professional IT experience, specializing in Machine Learning algorithms such as Ensemble
Methods (Random Forests), Data Analytics, Predictive Modeling, Natural Language Processing (NLP), Deep Learning
algorithms, and various forms of Regression (Linear, Polynomial, Logistic).
Proficient in Support Vector Machines (SVM), Deep Neural Networks, Extreme Gradient Boosting, Decision Trees, K-
Means, K-NN, Gaussian Mixture Models, Naive Bayes, and Convolutional Neural Networks (CNN), including LSTM
models. Expertise in implementing Time Series Models using RNNs, LSTMs, MMM, ARIMA, and SARIMA.
Data Scientist with a strong grasp of the Data Science lifecycle, OOP principles, Python , Streamlit, Fastapi,expertise,
and solid mathematics and statistics foundation.
Skilled in managing the entire data science project lifecycle: Data Cleaning, Exploratory Data Analysis (EDA), Data
Extraction, Data Pre-Processing, Feature Engineering, Dimensionality Reduction, Algorithm Implementation, Back
Testing, and Validation.
Competent in handling Structured and Unstructured data, Time Series data, and statistical methodologies like
Hypothesis Testing, ANOVA, multivariate statistics, modeling, decision theory, time-series analysis, and Descriptive
statistics. Utilized Azure Cognitive Services for advanced text analytics.
Leveraged JavaScript and Django for developing dynamic, data-driven web applications, integrating advanced
analytics and interactive features with a focus on scalable, efficient solutions.
Utilized GCP tools like BigQuery and Pub/Sub for managing large-scale data analysis and real-time processing,
optimizing data workflows and deploying scalable machine learning models.
Proficient in implementing Retrieval-Augmented Generation (RAG) to enhance natural language processing models
by integrating real-time retrieval mechanisms, improving the accuracy and contextual relevance of generative AI
outputs. Experienced with tools and frameworks like Hugging Face Transformers, ElasticSearch, Faiss, and OpenAI’s
API for implementing RAG workflows.
Skilled in working with Large Language Models (LLMs) for tasks such as text generation, summarization, and
translation, utilizing models like GPT, BERT, T5, and others within GCP’s Vertex AI, Azure OpenAI Service, and
Hugging Face's model hub. Proficient in using TensorFlow, PyTorch, and Transformers library for fine-tuning and
deploying these models at scale.
Experienced with Amazon Web Services (AWS), GCP, Azure, and Databricks, including cloud services like EMR,
EC2, S3, Vertex AI roles, and ETL processes using Hadoop, RDDs, DataFrames, MapReduce, and Spark-SQL.
Proficient in Artificial Intelligence, Data Science, Deep/Machine Learning Architect, Cognitive Services (Speech and
Voice Recognition), Statistics, Data Analysis using Python and MS Azure Machine Learning, Google Cloud Platform
and AWS.
have a strong background in building production-quality Retrieval-Augmented Generation (RAG) systems, configuring
vector databases, and using LangChain with sentence transformers. I am proficient in integrating advanced models
like GPT-4 and managing the MLOps lifecycle.
Expertise in Python, SQL, AI/ML cloud services (AWS, Azure), and hands-on experience with Metaflow, Flyte, and
Kubeflow for scaling machine learning workflows, with strong skills in communicating complex concepts to non-
technical audiences.
Strong background in Statistics, Probability, Linear Algebra, and Geometry.
Proficient in Python, R, SQL, PySpark, SAS; experienced with libraries and tools including Scikit-learn, NumPy, SciPy,
Pandas, Keras, TensorFlow, NLTK, and ggplot2.
Skilled in text mining, sentiment analysis, and brand reputation monitoring; knowledgeable in machine learning
applications across computer vision, predictive analytics, risk analytics, health monitoring, and IoT.
Familiar with R Studio (Matrices, Data Frames, Lists), Git for CI/CD, and Python frameworks for data processing and
model development.

TECHNICAL SKILLS:
 Languages: Python, R/R Studio, XML, SQL, Spark, pySpark, SparkSQL, Hadoop,
MapReduce,Node, NodeJS , JavaScript
(Packages: Stats, Matrix, data table, OpenSSL), Anaconda.
Statistics: Hypothetical Testing, MMM, Confidence Intervals, Bayes Law, Principal
Component Analysis (PCA), Cross-Validation, correlation.
Algorithms: Logistic Regression, Lasso Regression, XG Boost, KNN, SVM, Linear Regression,
Random Forest, Teradata, Generalized Linear Models,
AI, K-Means, SVN, Clustering, PuTTY, Neural Networks, AI Tableau, GitHub.
BI Tools: Tableau, Tableau server, Tableau Reader, SAP Business Intelligence, Amazon
Redshift, AWS Glue, Apache Spark,Azure Data Warehouse, Azure Data Factory,
Power BI, SAP
Business Objects.
Data Analysis Deep neural network, Logistic regression, Decision Tress, Random Forests, KNN,
and XG Boost, Ensembles (Bagging, Boosting), NLP, CNN, RNN, Recommendation
Data Science: models, Hadoop including Hive, HDFS, MapReduce and Spark, Spark SQL, Azure
ML, Scikit-learn, TensorFlow,Django, SpaCy, PyTorch, AutoML,
,LLM,Support Vector
Machines, Neural Networks, graph/network analysis, and Vertex AI,time series
analysis
(ARIMA model, FBProphet).
Reporting and Tableau, Power BI, Quick Sight, SAS Visual Analytics, IBM DataStage, Data Bricks.
ETL Tools:
Database Star Schema/Snowflake Schema modeling, Fact & Dimensions tables, physical
Design & logical data modeling, Normalization and De-normalization techniques,
Tools and Kimball& Inmon Methodologies
Data
Modeling:
EDUCATION:
Vignan’s Foundation for Science, Technology and Research- (Bachelor’s Degree)

PROFESSIONAL EXPERIENCE:

JP Morgan Chase, NYC Dec 2021- Present

ChatBot Developer/AI/ML Engineer
Responsibilities:
 Designed, implemented, deployed, and maintained deep learning and ML models using cloud technologies
such as Vertex AI, Google Storage, GCP Composer, and BigQuery.
 Automated data processing and model development using digital follow, MLflow and Apache Airflow, and
deployed machine learning models via Amazon SageMaker, AWS Lambda, and AWS Glue,improving the
efficiency and accuracy of model building.
 Developed and implemented deep learning models, including convolutional neural networks (CNNs) and
recurrent neural networks (RNNs), to solve complex problems in image recognition, natural language
processing, and sequence generation. Utilized advanced Python programming with Scikit-learn and NumPy
libraries.
 Utilized Spark, SQL, PySpark, Data Lake, TensorFlow, MLLib, and Python for various machine learning
methods, including classification, regression, recommendation, and dimensionality reduction, with experience
in Metaflow, Flyte, and Kubeflow for managing and scaling workflows.
 Wrote complex SQL statements to interact with RDBMS databases for data filtering and analytics.
 Developed reports, charts, tables, and other visual aids using Power BI and Tableau to support findings and
recommend business directions or outcomes.
 Wrote complex SQL queries to extract, transform, and analyze large datasets, ensuring high-quality data
pipelines.
 Integrated computer vision solutions with cloud-based MLOps pipelines using Kubernetes, MLflow, and AWS
Lambda for scalable deployment.
 Designed and deployed an AI-powered IVR system using Twilio Studio, Twilio Autopilot, and Azure Cognitive
Services, reducing call center dependency by 40%.
 Designed and deployed AI/ML models for financial risk assessment, fraud detection, and credit scoring using
TensorFlow, PyTorch, and Scikit-learn.
 Integrated image classification and NLP models to analyze handwritten financial documents and contracts for
compliance risk evaluation.
 Designed and fine-tuned GPT/LLM models with advanced prompt engineering and optimization techniques.
 Developed and optimized AI/ML models using PyTorch, TensorFlow, and R for predictive analytics.
 Optimized Gaussian Filtering for noise reduction in image processing and point cloud smoothing.
 Designed and deployed complex RAG pipelines without relying on LangChain or LlamaIndex, ensuring full-
stack implementation.
 Developed NLP models using SpaCy, Hugging Face, and NLTK for sentiment analysis, document processing,
and chatbot automation.
 Developed and fine-tuned NLP models using Azure Conversational Language Understanding (CLU) and
Twilio Voice to improve intent detection and sentiment analysis.
 Developed and deployed object detection, image segmentation, and classification models for real-time
inventory tracking using Faster R-CNN, and Mask R-CNN.
 Built image-based fraud detection models leveraging anomaly detection, GANs, and CNNs to identify
tampered checks, forged signatures, and document alterations in banking transactions.
 Developed automated check-processing models with image segmentation and handwriting recognition to
enhance banking workflows.
 Developed and deployed conversational AI chatbots using AWS Lex, integrating with Lambda for dynamic
response generation.
 Developed NER models using spaCy and transformers to identify and classify entities such as names, dates,
and locations in text data.
 Architected and implemented scalable MLOps pipelines leveraging AWS services such as SageMaker for
model training and deployment, Lambda for serverless model inference, and Step Functions for orchestrating
end-to-end workflows, ensuring streamlined model deployment across the organization.
 Optimized data pipelines and model monitoring by integrating S3 for data storage, CloudWatch for real-time
performance tracking, and Glue for automated data ETL, while enforcing role-based access control using IAM
and automating infrastructure provisioning with CloudFormation.
 Implemented custom NER pipelines using BERT, RoBERTa, and other transformer models for domain-
specific entity extraction.
 Leveraged foundation and open-source LLMs for scalable AI workflows, integrating vector databases like
Pinecone for efficient data retrieval.
 Implemented MLOps pipelines using MLflow and Kubeflow for model deployment and monitoring.
 Designed, deployed, and managed Kubernetes clusters for container orchestration and scaling.
 Designed, developed, and optimized RESTful APIs for seamless data ingestion and integration, enabling
efficient communication between data pipelines and various systems.
 Developed and optimized NLP applications using LangChain, enhancing model performance through effective
prompt engineering, chaining, and tool integration for real-time data retrieval and improved user interactions.
 Engineered and optimized radar data processing pipelines using Python, Apache Kafka, and SQL, ensuring
real-time integration with financial systems. This enhanced the accuracy of target detection and classification,
directly supporting fraud detection and risk management initiatives.
 Developed advanced machine learning models using TensorFlow and PySpark to analyze radar data,
providing actionable insights that informed strategic decisions in portfolio risk assessment and market
anomaly detection.
 Built an event-driven architecture using Twilio Webhooks, Azure Functions, and Cosmos DB to store and
analyze live call logs for customer behavior insights.
 Implemented OCR pipelines leveraging Tesseract, AWS Textract, and Google Vision API for automated
document verification in banking workflows.
 Developed real-time AI monitoring frameworks to track model drift and fraud pattern evolution in financial
datasets.
 Managed Public Key Infrastructure (PKI) in Azure, ensuring secure communication and handling of certificates
for PaaS services and applications.
 Applied a thorough understanding of the Data Science lifecycle and Object-Oriented Programming (OOP) to
build and deploy machine learning models using Python, TensorFlow, and PySpark.
 Leveraged mathematical and statistical methods, such as regression analysis (Linear, Logistic), hypothesis
testing (t-tests, Chi-Square), time-series forecasting (ARIMA, SARIMA), and advanced methods (Principal
Component Analysis, K-Means Clustering), along with tools like NumPy, Scikit-learn, and Spark for
comprehensive data analysis and financial risk assessment.
 Applied Gaussian Processes for uncertainty quantification in AI models.
 Implemented CI/CD pipelines, used Terraform for Infrastructure as Code, and developed automated testing
and monitoring solutions to enhance deployment efficiency and system reliability at JPMorgan.
 Implemented Kubernetes security best practices, including RBAC, network policies, and secrets management.
 Managed code repositories, branching strategies, and CI/CD workflows using GitHub.
 Integrated vector databases with AI workflows to enable retrieval-augmented generation (RAG) for improved
contextual responses.
 Developed custom PyTorch datasets, dataloaders, and augmentations for deep learning pipelines.
 Built and optimized AI-driven platforms using Azure, MongoDB for enhanced relevance.
 Designed and debugged Python applications in PyCharm, leveraging advanced development and debugging
tools for efficient coding.
 Managed Public Key Infrastructure (PKI) in Azure, ensuring secure communication and handling of certificates
for PaaS services and applications.
  Engineered and optimized cloud infrastructure by leveraging Azure Python SDK, Terraform, and GitHub
Runners to automate provisioning and deployment pipelines, ensuring seamless scalability and operational
efficiency.
 Designed and implemented vector search pipelines using databases like Pinecone for real-time semantic
search.
 Designed NLP-based virtual assistants with AWS Lex, leveraging intent recognition and slot filling for user
interactions.
 Configured and maintained Jenkins pipelines for automated builds, testing, and deployment.
 I have utilized Logistics, inferential models and data visualization to streamline operational processes,
enhance decision-making, and improve cost efficiency across business operations.
 Applied NLP techniques with tools like NLTK, Spacy, BERT, and SBERT models to automate and improve the
efficiency of JPMorgan’s customer service chatbots and internal document processing systems.
 Implemented and optimized vector databases such as FAISS, Milvus, and Pinecone for rapid and efficient
retrieval of financial records, improving data management processes.
 Developed and deployed Data Engineering APIs for model inference using frameworks like Flask, Django,
and FastAPI, facilitating seamless integration of machine learning models into production environments.
 Troubleshot and optimized Python code for machine learning models, ensuring reliability and performance in
production environments.
 Created and maintained detailed documentation for Python scripts, SQL queries, and machine learning
workflows to enhance project transparency.
 Developed and consumed APIs, Webhooks, and SDKs using Python, Git Actions, and Terraform, streamlining
cloud orchestration and integration with enterprise systems.
 Troubleshot complex Azure issues, including DNS, connectivity, NSG, and routing problems, ensuring reliable
and efficient cloud services.
 Deployed real-time analytics dashboards using Power BI and Azure Monitor, tracking IVR success rates,
customer sentiment, and drop-off points.
 Designed and implemented robust Cloud Infrastructure as Code (IaC) solutions, utilizing Terraform and
GitHub Actions, resulting in reduced manual intervention and enhanced CI/CD pipeline performance.
 Integrated data pipelines into modern microservices architecture, enabling seamless data streaming and
event-driven processing with Kafka.
 Enhanced the quality of insights by fine-tuning analytical models on curated financial and time-series datasets
to improve decision-making processes.
 Developed scripts for automated fine-tuning of models using tools like TensorFlow, PyTorch, and Vertex AI,
improving operational efficiency.
 Implemented best practices to enhance API performance, ensuring scalability and reliability in handling large-
scale data processing and real-time data streaming.
 Automated data workflows by integrating data Engineering APIs with tools like Apache Airflow and Apache
Kafka, streamlining data transformation, monitoring, and orchestration processes.
 Leveraged AI/ML cloud services (AWS SageMaker, Azure, Databricks) to automate financial models and
reduce operational costs. Utilized Git for version control and collaboration on SageMaker and other ML model
code. Presented ML insights to financial stakeholders, ensuring alignment with business goals and
maximizing platform efficiency.
 Designed processes for data quality validation and curation, addressing missing values, outliers, and
anomalies in both structured and unstructured datasets.
 Implemented a Generative AI solution using Gen AI tools and IBM Watson, OpenAI's Codex for generating
synthetic financial reports and detecting anomalies in transaction data.
 Validated AI data models using the K-fold technique to address overfitting and underfitting issues in deep
learning AI models.
 Designed and built ETL pipelines to automate the ingestion of structured and unstructured data.
 Actively participated in team meetings and discussions with business teams to understand use cases,
applying clustering techniques such as K-means, Gaussian mixture, and DBSCAN for customer
segmentation, product pricing, and customized marketing campaigns.
 Engineered Ensemble Models (e.g., Bootstrap Aggregating, Bagged Decision Trees, Random Forest,
XGBoost, AdaBoost), enhancing model accuracy by 40%, reducing bias/variance by 30%, and automating
processes for a 50% efficiency gain.

AIG, NYC, NY Jan 2021 – Dec 2021

AI/ML Engineer/ Data Scientist
Responsibilities:
 Developed and optimized an ML-based model using logistic regression and decision trees to classify
addresses as residential or commercial. Enhanced the model to determine business hours and customer
availability, resulting in a 6% reduction in delivery failures and a 15% regional improvement.
 Created a rule-based model with a 95% confidence interval and a 5-9% margin of error, adaptable across
regions, improving risk assessment and policyholder classification accuracy.
 Expertise in writing complex SQL queries for RDBMS and Spark SQL, utilizing cloud computing with
BigQuery, AWS (EC2, S3, Lambda, IAM, CloudWatch), and developing data ingestion pipelines using AWS
 Glue and Git workflows for seamless data integration.
 Utilized NER to improve data mining and text analytics for sentiment analysis, document classification, and
information retrieval.
 Debugged complex Python code using PyCharm's breakpoints, step-through execution, and variable
inspection tools to identify and resolve issues quickly.
 Built and enhanced media mix models, employing linear regression and random forest algorithms to link
marketing drivers to business outcomes, optimizing customer acquisition and policy sales.
 Implemented robust cloud infrastructure for data processing and ML model deployment using S3 for large-
scale data storage, IAM for role-based security, and CloudWatch for proactive system monitoring, improving
operational efficiency and reducing downtime in critical data workflows.
 Leveraged microservices to decouple complex ETL pipelines, improving maintainability and operational
efficiency in a multi-cloud environment.
 Collaborated with stakeholders to create entity annotation guidelines and ensure high-quality labeled datasets
for training and evaluation.
 Optimized vector database queries for high-speed retrieval and ranking of embeddings in large-scale
datasets.
 Optimized chatbot performance by fine-tuning AWS Lex models and integrating with DynamoDB for
contextual memory.
 Enhanced entity recognition by fine-tuning pre-trained models with labeled datasets to improve accuracy in
business-specific applications.
 Applied Graph Neural Networks (GNN) for AI-driven applications and knowledge graph solutions.
 Used PyTorch to develop deep learning models for computer vision, including CNNs for image classification
and object detection.
 Led the development of computer vision models for fraud detection, identifying inconsistencies in submitted
images for claims to reduce false claims.
 Wrote and maintained Azure Resource Manager (ARM) and Terraform templates, automating infrastructure
provisioning and improving resource management on Azure.
 Built end-to-end machine learning workflows, including data preprocessing, feature engineering, model
training, and evaluation.
 Built AI-powered virtual assistants and chatbots using Dialogflow CX, Google Assistant SDK, and RESTful
APIs for enhanced customer engagement.
 Configured and optimized PyCharm IDE for seamless integration with libraries, frameworks, and virtual
environments, enhancing productivity.
 Leveraged AWS SageMaker to build and deploy scalable machine learning models; applied PEFT and LoRA
techniques for optimizing large language models (LLMs) while minimizing compute overhead.
 Implemented Caffe, Theano-based models to detect anomalies in insurance transactions, enhancing fraud
detection and improving the accuracy of AIG’s financial security systems.
 Designed and deployed Data Engineering APIs , Streamlit, Fastapi , Django incorporating advanced ML
techniques, including support vector machines (SVM), K-means clustering, text analytics, and time series
forecasting, improving risk assessment, underwriting, and claims management.
 Fine-tuned machine learning models, including LLMs and recommendation systems, to optimize prediction
accuracy and performance on real-world datasets.
 Converted and integrated Jupyter Notebook code into production-ready Python scripts, ensuring scalability
and maintainability.
 Developed computer vision models for automated damage detection and claim assessment in insurance,
improving claim accuracy and processing speed.
 Deployed and maintained microservices using Kubernetes with Dockerized applications.
 Tracked development tasks, sprints, and issue resolution using JIRA and documented processes in
Confluence.
 Collaborated with cross-functional teams to iteratively refine IVR flows, improving NLP accuracy and reducing
misrouted calls by 25%.
 Designed and managed scalable cloud-based solutions on Google Cloud Platform (GCP) for high-
performance data processing.
 Deployed and managed Azure AI services for secure and scalable LLM solutions, leveraging Azure ML and
cognitive services.
 Implemented robust Azure DevOps pipelines to automate CI/CD for AI and data applications.
 Automated repetitive tasks and optimized performance by creating reusable Python modules for machine
learning and data processing.
 Engineered and optimized RESTful APIs and webhooks to integrate insurance applications with internal and
third-party services.
 Applied transfer learning and hyperparameter optimization techniques to adapt preazure-trained models for
domain-specific tasks, reducing training time and computational costs.
 Designed and implemented an AI-driven chatbot using Node.js and JavaScript, tailored to Insurance Services,
enhancing customer support by automating inquiries and providing real-time assistance through secure and
scalable solutions.
 Implemented image segmentation and object detection techniques to evaluate vehicle and property damage
 in real-time, reducing manual claim reviews by 30%.
 Integrated image segmentation models for automated property damage assessments, streamlining property
claims handling in real-time.
 Integrated NER systems into data processing pipelines for automated tagging and classification of large
volumes of unstructured data.
 Applied PyTorch’s DataLoader and Dataset classes to efficiently handle large datasets and implement data
augmentation techniques.
 Implemented Azure security features, including data protection, authentication, and RBAC, to enforce
compliance and secure cloud environments.
 Developed and automated data processing pipelines in JupyterLab using PySpark and Pandas for large-scale
financial datasets, enabling efficient data transformation and analysis for AIG's insurance business.
 Adopted an AIOps/MLOps mindset, developing automated pipelines to manage training, deployment, and
scaling of machine learning models in production environments.
 Utilized expertise in the Data Science lifecycle and Object-Oriented Programming (OOP) to develop advanced
predictive models and data pipelines with Python, R, and SQL.
 Applied mathematical and statistical methods, including regression analysis (Linear, Polynomial), probability
distributions (Normal, Binomial), statistical inference (Confidence Intervals, ANOVA), and time-series analysis
(ARIMA, Exponential Smoothing), using tools like SAS, Tableau, and Excel to improve risk management,
claims accuracy, and fraud detection.
 Automated cloud infrastructure using Azure CLI, enabling streamlined operations and faster response times
for cloud resource management.
 Monitored and optimized Vercel deployments, ensuring high availability and scalability for production
workloads.
 Designed and optimized MongoDB schemas to support AI applications requiring dynamic and scalable data
storage.
 Delivered Cloud Infrastructure as Code (IaC) solutions using Terraform and GitHub Actions, improving
deployment efficiency and reliability across cloud environments.
 Leveraged NLP libraries like NLTK, Spacy, and advanced models such as BERT and SBERT to automate
claims processing and enhance customer interaction, reducing operational costs.
 Advanced Radar Data Processing: Designed and implemented sophisticated radar data pipelines utilizing
Apache Spark, SQL, and MATLAB, enabling AIG to enhance real-time risk detection and classification. This
integration resulted in a 25% increase in predictive accuracy for insurance risk assessment.
 Comprehensive Data Analysis & Model Development: Conducted extensive exploratory data analysis (EDA)
on radar datasets using Python and R. Developed predictive models that seamlessly integrated with AIG’s
insurance systems, improving decision-making processes in underwriting and claims management.
 Integrated vector databases like Milvus, FAISS, and Pinecone to streamline data retrieval and processing for
underwriting and actuarial calculations, improving efficiency and accuracy.
 Ensured deployment and maintenance of AI-driven solutions in cloud environments (Google Cloud, AWS
SageMaker, Azure, Data Wrangler, and Databricks), adhering to AIG’s security and compliance protocols.
 Integrated advanced NLP and ML techniques into fraud detection and compliance systems, utilizing
frameworks like TensorFlow and PyTorch, ensuring adherence to financial regulations and improving
decision-making processes across various banking functions.
 Managed all phases of projects, including data collection, algorithm development, model validation, and report
generation, ensuring alignment with business goals in underwriting, claims processing, and fraud detection.
 Conducted Exploratory Data Analysis (EDA) using Pandas, NumPy, SciPy, and Jupyter Notebook. Created
dashboards with QuickSight, Tableau, and pivot tables, providing insights into policy performance and claims
trends.
 Managed Public Key Infrastructure (PKI) for Azure PaaS services, ensuring secure communication and
certificate management for cloud-based applications.
 Developed APIs and Webhooks using Python, Git Actions, and Terraform, streamlining cloud orchestration
and enabling multi-platform integrations for enterprise systems.
 Deployed machine learning models and data pipelines on AWS(Including Bedrock), leveraging services like
EC2, IAM, Lambda, and Serverless computing, ensuring scalable and efficient deployment of insurance
applications.
 Developed ML-based time series models using ARIMA, LSTM, and Prophet, leveraging cloud services and
parallelization techniques to optimize insurance claims forecasting and premium pricing.
 Utilized K-means clustering for customer segmentation based on demographic, geographic, and behavioral
data, optimizing policy offerings and marketing strategies. Automated Weekly Business Reviews (WBRs) with
detailed reports and visualizations using Excel and QuickSight

HPE, NYC, NY Feb 2019– Jan 2021

AI/ML Scientist/Data Scientist
Responsibilities:
 Identified, gathered, and preprocessed large-scale, multi-dimensional datasets using HPE's data management
systems, building efficient data pipelines for seamless integration and processing.
 Conducted in-depth data analysis with Python, Streamlit, Fastapi, R, and SQL, extracting actionable insights
 to drive data-driven decision-making across various business units.
 Utilized AWS and HPE's cloud platforms (EMR, S3, Apache Spark) for large-scale data processing and
analysis, ensuring high efficiency and scalability in handling complex datasets.
 Leveraged NLP libraries like NLTK, Spacy, and advanced models such as BERT and SBERT to automate
claims processing and enhance customer interaction, reducing operational costs.
 Integrated vector databases like Milvus, FAISS, and Pinecone to streamline data retrieval and processing for
underwriting and actuarial calculations, improving efficiency and accuracy.
 Developed and optimized machine learning models using TensorFlow, PyTorch, and HPE's AI/ML tools,
focusing on computer vision, natural language processing (NLP), and predictive analytics.
 Specialized in designing deep learning models for tasks like image classification, object detection, and
language translation, optimizing performance with hyperparameter tuning, model pruning, and transfer
learning.
 Leveraged Azure security features, including data protection, authentication, and RBAC, to ensure secure,
compliant cloud operations and protect sensitive resources.
 Implemented regression models including Linear Regression, Ridge Regression, and Lasso Regression for
predicting continuous outcomes.
 End-to-End Radar Data Integration: Built and optimized radar data processing pipelines using Apache
Hadoop, SQL, and Python, facilitating seamless integration with HPE’s edge computing platforms. This
enhanced HPE’s ability to deliver real-time, data-driven solutions for industrial automation.
 Predictive Analytics for Industrial Applications: Collaborated with DSP and ML engineers to develop and
deploy predictive models using TensorFlow, PyTorch, and Apache Spark. These models leveraged radar data
to improve HPE’s industrial automation systems, driving more accurate and efficient operational outcomes.
 Applied NLP models using NLTK, Spacy, and transformer architectures (BERT, SBERT) to enhance HPE’s
customer support systems and automate document processing.
 Developed Generative AI models for data augmentation in IoT and industrial automation applications,
leveraging LangChain and EmbedChain to enhance real-time processing capabilities.
 Deployed AI/ML models across hybrid cloud environments (Google Cloud, AWS, Azure), ensuring alignment
with HPE’s edge computing infrastructure and security standards.
 Designed and applied classification models such as Logistic Regression, Support Vector Machines (SVM),
Random Forest, Gradient Boosting Machines (GBM), and XGBoost for categorical predictions..
 Implemented end-to-end AutoML solutions and designed ML pipelines using Vertex AI and HPE's AI
platforms, automating model training, evaluation, and deployment processes.
 Applied AI/ML algorithms to edge computing scenarios, deploying models on HPE's edge devices for real-time
data processing and decision-making, particularly in IoT and industrial automation applications.
 Developed dynamic and interactive dashboards using Tableau and Power BI, presenting complex data and
model outcomes to stakeholders with clarity and precision.
 Ensured AI models adhered to ethical guidelines, emphasizing fairness, transparency, and accountability, and
aligned with industry standards and HPE's compliance requirements.
 Collaborated with cross-functional teams to develop and deploy AI solutions, utilizing Git for version control
and continuous integration/continuous deployment (CI/CD) pipelines.

Accenture, India May 2015- July 2016

Data Engineer
Responsibilities:
 Optimized Spark applications by tuning batch intervals, parallelism, and memory allocation to ensure efficient
processing of healthcare data, adhering to HIPAA regulations.
 Applied triage techniques to assess, prioritize, and address critical issues promptly, driving process
improvements within the triage workflow.
 Led the migration of on-premises data to Snowflake, optimizing performance, scalability, and cost-efficiency
with comprehensive data models.
 Stayed updated with Snowflake features and best practices, contributing to system optimizations and
operational excellence.
 Utilized Spark Core, Spark SQL, and Scala for data processing, and developed Python APIs for debugging
and monitoring.
 Managed Spark cluster size, monitoring, and troubleshooting, and implemented triage processes for efficient
issue management.
 Developed Spark applications with PySpark and Spark-SQL for data extraction and transformation, and
designed data models for enterprise data lakes.
 Leveraged Microsoft Azure components, including Azure Data Factory, Data Lake Storage, and Azure
Databricks for data processing and analytics.