SUMMARY:

• AI/ML Engineer (8+ years) delivering end-to-end intelligent systems with Machine Learning, Deep Learning, NLP, and Agentic AI
• Expert in designing and deploying LLM-powered AI agents for workflow automation, dynamic reasoning, and real-time decision support across enterprise domains.
• Skilled in LLMs, function/tool calling, multi-agent orchestration, and RAG using LangChain, LangGraph, and Semantic Kernel.
• Proficient in Python, PyTorch, TensorFlow, Scikit-learn, Pandas, NumPy, and PySpark with strong foundations in ML algorithms, predictive modeling, and data analytics.
• Hands-on with MLOps workflows – Azure ML, MLflow, Kubeflow, Seldon Core, and CI/CD for continuous integration, model versioning, and production deployment.
• Experienced in containerization and cloud-native deployment using Docker, Kubernetes (AKS/EKS/GKE), and terraform for scalable, resilient AI infrastructures.
• Built and managed voice & multimodal agents integrating ASR, TTS, LLM reasoning, and API tools, enhancing automation and user interaction in production workflows.
• Designed RAG pipelines leveraging vector databases (pgvector, FAISS, Qdrant) for grounded, factual AI responses.
• Strong grasp of prompt engineering, fine-tuning (LoRA/QLoRA, RLHF), and model optimization for efficiency and interpretability.
• Experienced in AI safety & observability using OpenTelemetry, Prometheus, and Grafana
• Adept at connecting AI agents to enterprise systems and APIs, automating scheduling, data entry, and operational workflows.
• Implemented secure, compliant MLOps pipelines with focus on transparency, fairness, and ethical AI.
• Proficient in the data science lifecycle – data ingestion, EDA, feature engineering, model training, and post-deployment monitoring.
• Built and optimized LLM-based conversational and reasoning agents for healthcare, analytics, and automation
• Hands-on with Azure, AWS, GCP, and Databricks for model serving, data engineering, and distributed training.
• Skilled in text analytics & NLP: tokenization, embeddings, sentiment analysis, topic modeling, and information retrieval.
• Strong background in statistical modeling, hypothesis testing (ANOVA, t-tests, Chi-Square), and regression/classification techniques for insight-driven decision support.
• Experienced in reinforcement learning and adaptive agents to improve long-term system performance.
• Collaborated with cross-functional teams to integrate AI solutions into production systems via REST APIs and event-driven architectures.
• Delivered enterprise-grade AI solutions, including a multi-agent CDC chatbot using Azure OpenAI GPT, RAG pipelines, and secure dual-agent routing for internet and intranet data retrieval.
• Developed and deployed a healthcare AI system for automated pain-pattern detection using Azure ML, OpenCV, and TensorFlow – transforming scanned BPI body maps into structured clinical datasets for predictive research.
• Passionate team player dedicated to developing responsible AI that delivers real-world value and operational excellence.

CERTIFICATIONS:

• Microsoft Certified: Azure AI Engineer Associate
• Microsoft Certified: Azure Data Scientist Associate
• NVIDIA Certified Deep Learning Institute (DLI)
• AWS Certified Machine Learning – Specialty

EDUCATION:

• Fairfax University of America, VA

Master of Science: Computer Science

• HU University of Science and Technology, PA

Master of Science: Data Science

TECHNICAL SKILLS:

Machine Learning & Statistical Modeling	Supervised and Unsupervised Learning: Linear/Logistic regression, Decision trees, SVMs, KNN, Random Forests, XGBoost, Clustering Feature Engineering: Dimensionality reduction (PCA, LDA, SFS/SBS/SFFS) Statistical Testing: Anova, t-tests, Chi-Square Time Series Forecasting: ARIMA, SARIMA, Prophet Ensemble Learning: Random Forests, Gradient Boosting Recommendation engine: Market Basket Analysis, Collaborative filtering, Segmentations
Deep Learning & Gen AI	Neural network design: ANN, CNN, RNN, Transformer architectures using PyTorch and TensorFlow. (LLM) engineering: prompt design, function/tool calling, LoRA/QLoRA fine-tuning, and retrieval-augmented generation (RAG) pipelines. Voice AI: Automatic Speech Recognition (ASR), Text-to-Speech (TTS), Voice Activity Detection (VAD), Barge-In, and Emotion/Intent detection. Multimodal agent: Whisper, Azure Speech, ElevenLabs, LangChain, LlamaIndex, and OpenAI function calling
Agent & Orchestration Frameworks	Conversational and autonomous agents: LangChain, OpenAI Agents, CrewAI, Semantic Kernel, and Azure AI Studio. State management, memory, and context caching via Redis, pgvector, and FAISS/Qdrant. Event-driven and streaming orchestration: FastAPI, WebSockets, gRPC, and Twilio/WebRTC integrations.
MLOps & Deployment	Azure ML, MLflow, Kubeflow, Seldon Core, Docker, Kubernetes (AKS), and IaC tools (Terraform, Azure ARM, CloudFormation), Prometheus, Grafana, and OpenTelemetry.
Cloud Platforms & Data Systems	Azure (OpenAI, Cognitive Services, AKS, Blob, Cosmos DB), AWS (SageMaker, Lambda), and GCP (Vertex AI, BigQuery), PySpark, Databricks, Snowflake, PostgreSQL, MySQL, MongoDB, Cassandra, Git, GitHub Actions, and CI/CD pipelines.
Data Engineering & Cloud Analytics	Azure Data Factory, AWS Glue, Apache Airflow, Kafka, Kinesis, and Event Hub

WORK EXPERIENCE:

Booz Allen Hamilton, GA Sep 2021– Present

Lead AI/ML Engineer

Responsibilities:

• Serve as Lead AI Scientist and Subject Matter Expert for Databricks, Azure Machine Learning, and R/Posit, driving AI and data initiatives across healthcare and federal divisions.
• Conduct research, prototyping, and architecture design for large-scale AI and MLOps implementations, guiding cross-functional teams in solution development and deployment.
• Designed and deployed end-to-end AI/ML solutions in Azure Machine Learning, leveraging registered data assets, versioned environments, and reproducible pipelines for healthcare analytics.
• Led the Automated Pain Pattern Detection from BPI Body Maps project, developing a full computer vision pipeline to extract and analyze patient pain markings using OpenCV, TensorFlow, and Azure ML pipelines.
• Created a 74-zone anatomical mapping system based on the CHOIR framework, enabling structured pixel-level pain localization for chronic pain research datasets.
• Implemented image alignment via ORB + RANSAC-based homography and pain-mark isolation using dual thresholding, morphological operations, and Gaussian blurring, producing high-quality binary pain masks for analysis.
• Developed algorithms for centroid detection, zone coverage, and circle-based segmentation, mapping patient pain marks to standardized body zones.
• Trained and deployed a CNN-based marking style classifier using TensorFlow/Keras on Azure ML Compute, automating image routing to correct extraction logic.
• Built custom Azure ML environments and automated retraining pipelines for reproducible experiments and dependency consistency across compute targets.
• Ensured HIPAA compliance and data security through encryption, PHI redaction, and Azure-controlled access policies for clinical datasets.
• Designed and developed an internal multi-agent chatbot for the Centers for Disease Control and Prevention (CDC) using Azure OpenAI GPT models, integrating both internet-enabled and intranet-restricted search capabilities.
• Architected a dual-agent framework to route queries securely between public web search and CDC confidential repositories, governed by role-based access control (RBAC) and audit logging.
• Implemented retrieval-augmented generation (RAG) using Azure Cognitive Search enabling contextual responses grounded in vetted CDC datasets and internal documents.
• Built embedding pipelines to index policy manuals, research archives, and structured health data for semantic retrieval and conversational reasoning.
• Deployed the chatbot in a containerized Azure Kubernetes Service (AKS) environment with FastAPI backend, Docker, and Active Directory
• Contributed to the development of an AI Voice Agent prototype integrated with the internal CDC chatbot, enabling speech-to-text (ASR), LLM-based understanding, and text-to-speech (TTS) for hands-free conversational interaction.
• Supported the design of streaming voice pipelines using Twilio/WebRTC ‚ASR ‚Azure OpenAI ‚TTS, ensuring low-latency response times and smooth bidirectional communication.
• Assisted in refining prompting logic, barge-in behavior, and function-calling schemas to improve natural conversation flow and enhance user experience within secure CDC environments.
• Collaborated with CDC cybersecurity and data governance teams to implement responsible AI guardrails, mitigate prompt injection risks, and ensure alignment with federal data privacy regulations.
• Created observability dashboards (Grafana, Application Insights, OpenTelemetry) to track latency, agent routing efficiency, and LLM response accuracy.
• Built CI/CD pipelines using Azure DevOps, Docker, and Jenkins for model versioning, automated testing, and continuous deployment of AI systems.
• Mentored junior data scientists, reviewed model code and architecture, and established standards for secure, ethical, and transparent AI

Environment: Azure Machine Learning, Azure OpenAI (GPT-4/GPT-3.5), Databricks, TensorFlow, Keras, OpenCV, LangChain, pgvector, Azure Cognitive Search, FastAPI, PySpark, Python (Pandas, NumPy, Scikit-learn), Power BI, Tableau, Docker, Kubernetes (AKS), Azure DevOps, Jenkins, Grafana, Application Insights.

RGP, GA Feb 2021– Aug 2021

Senior Consultant

Responsibilities:

• Led end-to-end data science lifecycle – data collection, cleaning, exploration, feature engineering, model training, validation, and insight delivery – to drive data-informed decisions.
• Collaborated with Data Engineers, Analysts, and Business stakeholders to translate analytical objectives into scalable predictive models and business impact.
• Conducted Exploratory Data Analysis (EDA) using Python (Pandas, NumPy, Matplotlib) to validate hypotheses, identify missing values, manage categorical variables, and handle data imbalance using resampling and SMOTE
• Performed feature engineering and dimensionality reduction (correlation analysis, Principal Component Analysis) to optimize model performance and interpretability.
• Utilized Spark, PySpark, and SparkML to process and model large datasets efficiently, ensuring scalability and high-performance training.
• Designed and deployed machine learning models using Scikit-learn, Spark MLlib, and TensorFlow, including Logistic Regression, Random Forests, Decision Trees, K-Nearest Neighbors (KNN), Gradient Boosting, and Neural Networks.
• Applied model evaluation metrics – confusion matrix, precision, recall, F1 score, ROC-AUC, and K-fold cross-validation – to measure model robustness and reliability.
• Conducted unsupervised learning using K-Means clustering to segment customers and uncover actionable behavioral patterns.
• Developed and optimized SQL queries for complex data mining, feature extraction, and profiling to support predictive analytics workflows.
• Created visual analytics and dashboards using Matplotlib, ggplot2, and Power BI, translating model outputs into clear business insights and recommendations.
• Leveraged Spark MLlib optimization algorithms for regression, classification, and clustering tasks at scale.
• Presented analytical findings and predictive insights through reports and stakeholder presentations, enabling data-driven strategic decisions.
• Demonstrated strong problem-solving, analytical reasoning, and communication skills in translating technical analysis into tangible business value.
• Conducted time-series forecasting to predict operational trends and resource utilization using models such as ARIMA, Prophet, and Gradient Boosted Trees.
• Engineered data validation and drift-detection checks within ML pipelines to ensure model reliability and trigger retraining when statistical thresholds were breached.
• Performed hyperparameter optimization using GridSearchCV and Bayesian Optimization to maximize model accuracy and reduce overfitting.
• Integrated MLflow for model tracking, experiment comparison, and version control, ensuring reproducible development across environments.
• Partnered with data engineering teams to establish ETL pipelines and implement data-quality rules that improved dataset reliability for analytics.
• Designed statistical control charts and anomaly-detection models to identify irregularities in production data streams.
• Collaborated with domain experts to define key performance indicators (KPIs) and translate model outputs into measurable business metrics.

Environment: Python (Pandas, NumPy, SciPy, Scikit-learn, Matplotlib), TensorFlow, Spark, PySpark, SparkML, SQL, Power BI, ggplot2 CGI, VA

Senior Consultant Jan 2017 – Aug 2021

Responsibilities:

• Served as a Data Specialist, Data Scientist, and Production Support Engineer, managing end-to-end data operations, database administration, and predictive analytics to ensure system reliability, optimize performance, and drive data-informed decision-making.
• Administered and optimized enterprise databases (AWS RDS, Redshift, DynamoDB), performing schema design, indexing, backup/recovery, and access-control management for high-availability systems.
• Delivered 24/7 production support for mission-critical data platforms, monitoring job execution, performing health checks, and implementing automated recovery mechanisms to reduce downtime.
• Developed predictive incident models to forecast system outages, support-ticket spikes, and query slowdowns, using machine-learning algorithms such as Random Forest, Logistic Regression, and Gradient Boosting.
• Built automated anomaly-detection pipelines to identify unusual resource usage patterns and alert operations teams before performance degradation occurred.
• Applied statistical modeling and root-cause analytics in R and Python, identifying trends, dependencies, and anomalies leading to data service disruptions.
• Designed and implemented ETL and data-ingestion frameworks integrating APIs, S3 buckets, and databases for consistent data synchronization across environments.
• Created data quality validation scripts in Python and SQL to flag missing, duplicate, and out-of-range data points, ensuring accurate and trusted analytics.
• Utilized PySpark, Spark SQL, and MLlib for large-scale data processing, feature extraction, and distributed model training on terabyte-scale datasets.
• Engineered ML-based root-cause analysis tools that reduced manual troubleshooting time and enabled early detection of recurring data incidents.
• Deployed machine learning models on AWS SageMaker and Lambda, integrating outputs into production alert systems to drive real-time decision support.
• Designed feature-engineering pipelines (normalization, scaling, PCA, encoding) using Scikit-learn, improving prediction accuracy and model stability.
• Implemented hyperparameter optimization (Grid Search, Random Search, K-Fold Cross Validation) to improve accuracy across classification and regression models.
• Developed automated retraining and model tracking workflows using MLflow, ensuring continuous improvement of predictive performance.
• Designed real-time data streaming workflows using AWS Kinesis and Spark Structured Streaming for proactive operational monitoring.
• Created interactive Tableau dashboards to visualize key performance indicators (KPIs), database load patterns, and predictive model outcomes for business and technical stakeholders.
• Collaborated with data-engineering, DevOps, and application teams to design self-healing data jobs, implement alert systems, and automate release processes through CI/CD
• Performed query optimization and stored procedure tuning in SQL and PL/SQL to enhance report generation and reduce execution time.
• Analyzed and forecasted resource utilization trends to plan compute scaling strategies and budget allocations across production environments.
• Ensured data security and compliance through encryption, IAM role enforcement, and regular access audits across all AWS services and on-prem databases.
• Implemented logging, telemetry, and monitoring frameworks (CloudWatch, ELK, OpenTelemetry) to ensure observability and traceability of all critical data flows.
• Built incident trend dashboards combining ML predictions and ticketing-system data to prioritize high-impact recurring issues.
• Partnered with infrastructure teams to perform cost-performance optimization of AWS clusters and EMR workloads, achieving measurable savings without impacting SLAs.
• Participated in disaster recovery and business continuity planning, validating backup integrity and failover readiness across multiple regions.
• Mentored junior engineers and analysts on SQL performance tuning, data modeling, and predictive-maintenance methodology.
• Recognized for combining operational reliability, analytical rigor, and proactive automation, significantly improving production efficiency and incident management outcomes.

Environment: AWS (S3, EMR, RDS, Redshift, DynamoDB, SageMaker, Lambda, CloudWatch, QuickSight, Kinesis), Python (Pandas, NumPy, SciPy, Scikit-learn, Matplotlib, Seaborn), R, Spark (PySpark, Spark SQL, MLlib), Airflow, Tableau, MLflow, SQL, GitHub, DevOps (CI/CD), OpenTelemetry, ELK Stack