Home

Achyutha Pranavi - Sr. AI/ML Engineer | Data Analyst

[email protected]

Location: , , USA

Relocation: REMOTE

Visa: GC

Resume file: AchyuthaPranavi_SeniorAI-ML_Engineer_DataAnalyst_Resume_2025_1755703281665.docx Please check the file(s) for viruses. Files are checked manually and then made available for download.

Name: ACHYUTHA PRANAVI K Role: Senior AI/ML Engineer | Senior Data Analyst Email address: [email protected] Contact: +1 (347) 921-0136 LinkedIn: http://www.linkedin.com/in/achyutha-p-89a3b5216 Professional Summary Results-driven Senior AI/ML Engineer and Senior Data Analyst with 11 years of experience in data science, machine learning, advanced analytics, SAS programming, and AI solutions across finance, healthcare, retail, insurance, and telecom sectors, delivering measurable business impact through cutting-edge implementations. Expert in cloud-native AI/ML solutions on Microsoft Azure (Azure ML, Databricks, Synapse Analytics, Cognitive Search, Azure OpenAI) and AWS (Redshift, S3, Lambda, Athena, Glue, QuickSight), GCP (BigQuery, Vertex AI, Dataflow), architecting and deploying scalable, production-ready ML/AI systems. Skilled in Python, TensorFlow, PyTorch, R, SQL, and cloud-based MLOps workflows (AWS SageMaker, Azure ML, Databricks) for developing, optimizing, and deploying ML/AI models, including real-time inference pipelines. Advanced practitioner in Large Language Models (LLMs), Generative AI, Cursor AI, and Agentic Workflows, building retrieval-augmented generation (RAG) pipelines and localization-focused AI solutions using LangChain, Semantic Kernel, FAISS, Azure Cognitive Search, LlamaIndex, Spring AI, and Agno, enhancing accuracy and reducing hallucinations. Specialized in building Agentic AI systems where LLM-powered agents autonomously plan, execute, and validate complex workflows across finance and healthcare domains. Earlier at HCA Healthcare, I pioneered agentic AI prototypes for compliance Q&A and claims fraud detection, laying the groundwork for multi-agent healthcare automation. Experienced in advanced AI architecture patterns including RAG, Agentic RAG, MCP (Model Context Protocol), Function Calling, and Agent-to-Agent (A2A) communication within Azure environments. Developed transformer-based NLP models (BERT, GPT, Azure OpenAI) for document summarization, sentiment analysis, entity/risk detection, and clinical note processing, enabling automated compliance and research workflows. Skilled in fine-tuning, reinforcement learning (RLHF), prompt-tuning, and post-processing to improve model alignment, accuracy, and relevance. Designed and deployed multi-agent systems using LangChain, LangGraph, CrewAI, AutoGen, and Cursor AI for intelligent automation in finance, healthcare, and compliance reporting. Applied time-series forecasting models (ARIMA, Prophet, LSTM, Neural Prophet) for financial trend prediction, demand forecasting, and risk assessment in lending, trading, and retail operations. Built comprehensive MLOps and CI/CD pipelines using MLflow, Azure DevOps, Docker, and Kubernetes, ensuring reproducibility, monitoring, governance, and compliance in high-stakes deployments. Engineered feature engineering and ETL pipelines with PySpark, pandas, Azure Data Factory, AWS Glue, Delta Lake, and Alteryx, delivering high-quality datasets that significantly improved model performance. Delivered enterprise analytics and BI solutions using Power BI, Tableau, and QuickSight for real-time KPI tracking and executive reporting across multiple industries. Implemented responsible AI practices with SHAP, LIME, fairness metrics, and Azure AI Content Safety, ensuring explainability, transparency, and regulatory compliance (HIPAA, financial services). Designed predictive models for fraud detection, claims management, and patient stratification, achieving 30% reduction in false claims and 18% improvement in fraud detection accuracy. Applied knowledge graph models with Neo4j, NetworkX, and graph embeddings to map complex relationships in fraud detection and compliance analytics. Conducted advanced statistical analysis, A/B testing, cohort analysis, and segmentation, contributing to 23% campaign targeting improvement and 12% revenue uplift. Built scalable data pipelines processing financial transactions, healthcare records, and retail data using Snowflake, Alteryx, and distributed computing frameworks. Leveraged self-supervised learning and contrastive learning for domain-specific pretraining on financial and healthcare text corpora, improving embedding quality for downstream ML tasks. Developed synthetic data generation workflows with Gretel.ai and internal tools to address imbalanced datasets while ensuring privacy, security, and compliance. Proven expertise in SQL query optimization and data warehousing using Amazon Redshift, Snowflake, SQL Server, MySQL, Oracle, delivering near real-time reporting and 30% system performance improvements. Conducted sales funnel, conversion rate, and clickstream data analysis using Python (pandas, NumPy) and R, identifying growth opportunities across 1,800+ retail locations. Integrated external datasets (e.g., Nielsen, social media sentiment, market data, social determinants of health) with internal data for enriched analytical models and competitive insights. Implemented data governance, lineage tracking, and metadata management using AWS Lake Formation, Unity Catalog, and dbt, ensuring standardization and regulatory compliance. Automated analytical workflows with Python scripting, AWS Lambda, and Azure Functions, reducing manual workload by 40% and enabling real-time insights. Experienced in private equity AI platform development and integration, leveraging Azure infrastructure, Flask-based API services, and secure microservices design. Delivered cross-functional leadership and mentorship, conducting knowledge-sharing sessions on MLOps, Generative AI, Azure AI Studio, and translating complex AI models into actionable business insights. Technical Skills Programming & Core Technologies Python, Java, JavaScript, SQL, SAS, PySpark, Bash, R, Flask, Pydantic, MSSQL, Linux/Unix Cloud Platforms Microsoft Azure (Azure ML, Azure Databricks, Azure Synapse, Azure Cognitive Search / Azure AI Search, Azure OpenAI, Azure ML Studio), AWS (SageMaker, Redshift, S3, Lambda, Athena, Glue, QuickSight), GCP (BigQuery, Vertex AI, Dataflow, Looker Studio), Snowflake, Databricks Data Visualization & BI Power BI, Tableau, Amazon QuickSight, Matplotlib, Seaborn Big Data & ETL PySpark, pandas, Azure Data Factory, AWS Glue, SQL, Alteryx, Delta Lake, Apache Airflow Databases & Warehousing Amazon Redshift, Snowflake, SQL Server, MySQL, Oracle, Data Modeling, Query Optimization Statistical Analysis & Business Intelligence Advanced Statistics, A/B Testing, Cohort Analysis, Customer Segmentation, Market Research, KPI Development, Forecasting & Demand Planning AI/ML Frameworks & Agentic AI & Advanced Learning PyTorch, TensorFlow, scikit-learn, Keras, XGBoost, Hugging Face Transformers Agentic AI (Core Expertise): LangChain, LangGraph, CrewAI, AutoGen, Cursor AI, Semantic Kernel, Spring AI, Agno Agentic AI Concepts: Multi-Agent Systems, A2A Communication, Function Calling, Model Context Protocol (MCP), Planner Executor Workflows Self-Supervised & Contrastive Learning, Prompt/Context Engineering, RAG Pipelines NLP & LLMs BERT, GPT, Azure OpenAI, RAG (FAISS, Vector DBs), Text Summarization, Sentiment Analysis, Entity Recognition, Localization QA, Multi-lingual NLP, Translation Consistency Checks Time-Series Forecasting ARIMA, Prophet, LSTM, Neural Prophet Vector & Graph DBs FAISS, Azure Cognitive Search / Azure AI Search, Neo4j, NetworkX, Graph Embeddings, Knowledge Graphs MLOps & DevOps MLflow, Azure DevOps, AWS SageMaker, Azure ML, Databricks, Docker, Kubernetes, Prometheus, Azure Monitor, CI/CD for ML, Real-Time Model Inference Pipelines, Unit Testing, Postman Responsible AI SHAP, LIME, Fairness Metrics, Azure AI Content Safety Synthetic Data & Privacy Gretel.ai, Differential Privacy, Data Augmentation Educational Details Master of Science in Computer Science - University of Central Missouri (Aug 2011 - Dec 2012) Bachelor of Science in Computer Science - Lovely Professional university (Aug 2007 - Jun 2011) Certifications Microsoft Certified: Azure AI Engineer Associate - 2023 AWS Certified Machine Learning - Specialty - 2021 Work Experience Client: Jefferies Financial Group Inc, New York, NY May 2024 - Present Role: Senior AI/ML Engineer Responsibilities: Designed and implemented AI/ML solutions on Microsoft Azure using Azure ML, Azure Databricks, and Azure Synapse Analytics for scalable, secure model training, deployment, and governance. Led end-to-end ML pipeline development with Python, Azure ML SDK, and MLflow, integrating CI/CD and governance for reproducibility and operational excellence. Built and deployed NLP/LLM models for financial document summarization, sentiment analysis, risk entity detection, and localization QA, leveraging BERT, GPT, Azure OpenAI, and custom transformers. Designed Agentic AI workflows that coordinated multiple autonomous agents (retriever, summarizer, validator) using LangChain, LangGraph, CrewAI, LlamaIndex, Spring AI, Agno, Cursor AI and AutoGen - enabling end-to-end financial research automation with minimal human oversight. Built multi-agent systems for compliance and earnings analysis, where agents collaborated to extract insights, validate against rules, and generate executive-ready reports cutting manual analyst workload by 65% and reducing errors by 30%. Implemented Agent-to-Agent (A2A) communication and function-calling orchestration, allowing LLM-powered agents to trigger APIs, query databases, and cross-verify outputs in real time for audit-proof financial reporting. Developed RAG pipelines with FAISS, Azure Cognitive Search (AI Search), applying embedding, chunking, grounding, prompt engineering, and context engineering strategies to reduce hallucinations and improve retrieval accuracy. Evaluated GCP Vertex AI and BigQuery alongside Azure ML to benchmark RAG pipeline portability and performance for financial document summarization, ensuring multi-cloud flexibility and cost optimization. Enhanced LLM/NLP performance through prompt tuning, PEFT (parameter-efficient fine-tuning), RLHF, and post-processing optimizations. Applied time-series forecasting models (ARIMA, Prophet, LSTM) for multi-asset financial trend prediction and risk forecasting in lending and trading use cases. Engineered feature pipelines using PySpark, pandas, Azure Data Factory, and Databricks Delta Lake, enabling advanced transformations and temporal joins. Applied knowledge graph modeling (Neo4j, NetworkX) to map customer transaction instrument relationships for fraud detection and compliance analytics. Developed REST APIs (Flask, Python) and hybrid Python-Java microservices integrated with AI APIs for financial analytics; validated using unit tests & Postman. Applied design patterns (MVC, Singleton, Factory) for scalable AI microservices and enforced application security (SSL/TLS, JWT, RBAC). Implemented real-time inference pipelines for high-frequency financial streams, reducing latency by 35% and enabling sub-second decision-making. Built synthetic data generation workflows with Gretel.ai and internal tools to augment imbalanced datasets while protecting sensitive financial data. Created monitoring dashboards with Power BI, Azure Application Insights, Prometheus to track model drift, prediction accuracy, and anomalies. Researched self-supervised and contrastive learning for pretraining on proprietary financial corpora to improve domain embeddings. Applied data mesh principles for scalable ML architectures across global financial domains, enabling decentralized innovation. Integrated Responsible AI practices with SHAP, LIME, and Azure AI Content Safety to ensure transparency, fairness, and compliance with financial regulations. Participated in Agile/Scrum ceremonies (Planning, Standups, Retrospectives), collaborating across engineering, compliance, and trading teams. Created executive reporting suites in Power BI and Azure Synapse Analytics, delivering real-time insights on trading volumes, market volatility, and portfolio performance. Mentored engineers and analysts, leading training on MLOps, Generative AI, Azure AI Studio, and foundation model adaptation, establishing best practices for production-scale AI in finance. Client: HCA Healthcare Inc, Nashville, TN Nov 2022 - April 2024 Role: AI/ML Engineer Responsibilities: Designed and deployed ML models for patient risk prediction and stratification, improving proactive care delivery and clinical decision support. Partnered with clinical and engineering teams to translate healthcare goals into data-driven ML solutions using Azure Machine Learning Studio, Azure Synapse Analytics, and Azure Databricks. Built ETL/data pipelines with Azure Data Factory, PySpark, and SQL, ensuring accurate ingestion and transformation of large-scale healthcare datasets. Developed fraud detection models for claims, achieving a 30% reduction in false claims using supervised ML. Created deep learning models with TensorFlow and PyTorch for classification, anomaly detection, and time-series forecasting in healthcare operations. Integrated explainable AI (SHAP, LIME) to ensure transparency, compliance, and clinician adoption in predictive models. Prototyped early Agentic AI workflows by combining LangChain, Azure OpenAI, and Azure Cognitive Search to create autonomous clinical Q&A assistants that retrieved compliance documents, answered physician queries, and flagged missing information. Designed multi-agent validation flows where one agent extracted patient data, another verified clinical coding accuracy, and a third checked HIPAA compliance reducing audit preparation time by 40%. Integrated RAG + agent-based orchestration to build intelligent assistants for claims fraud detection and medical note summarization, enabling self-checking workflows that improved accuracy and trust in clinical decision support systems. Leveraged Azure AutoML for hyperparameter tuning and accelerated model experimentation. Leveraged GCP Dataflow for real-time ingestion of clinical data streams in a POC, comparing performance and cost efficiency against Azure Data Factory. Implemented MLOps pipelines using Azure DevOps, MLflow, Docker, and Kubernetes (AKS) for reproducible and scalable deployments. Developed monitoring dashboards with Power BI and Azure Monitor for drift detection, accuracy tracking, and compliance reporting. Integrated social determinants of health (SDoH) datasets with internal records on Azure Data Lake, enhancing predictive models through feature engineering. Applied transfer learning to diagnostic imaging use cases, improving anomaly detection accuracy across medical scans. Supported annotation workflows and active learning for medical imaging and text datasets, reducing labeling overhead. Contributed to data privacy & HIPAA compliance, embedding security and governance controls across the ML lifecycle. Mentored junior engineers on Azure ML workflows, Git version control, and MLflow tracking. Delivered presentations to clinicians and leadership, demonstrating AI s role in value-based care and operational efficiency. Client: Target Corp, Minneapolis, MN Jan 2019 - Oct 2022 Role: Senior Data Analyst Responsibilities: Led the design and implementation of scalable AWS data pipelines using Amazon Redshift, S3, Glue, and Athena to process and store high-volume retail transaction data across 1,800+ locations. Delivered actionable insights with complex SQL queries, stored procedures, and SAS reports, improving reporting speed by 35% and enabling deeper customer behavior and product trend analysis. Created executive dashboards in Power BI, Tableau, and Amazon QuickSight, providing real-time visibility into inventory optimization, supply chain efficiency, and sales KPIs. Collaborated with marketing, merchandising, and e-commerce teams to align analytics with business goals, improving campaign targeting by 23%. Conducted sales funnel, conversion rate, and clickstream analysis using Python (pandas, NumPy), identifying friction points in the customer journey. Managed data modeling projects with dbt and Snowflake, ensuring consistency in dimensions/metrics and strengthening data integrity across teams. Designed ETL workflows in Alteryx to automate ingestion and transformation of financial and operational data. Leveraged Snowflake data warehousing to prepare model-ready datasets, laying groundwork for future RAG/LLM use cases in retail knowledge bases. Integrated GCP BigQuery in a hybrid retail analytics workflow to process promotional campaign data, later migrated insights into AWS Redshift for enterprise reporting. Applied advanced statistical models and demand forecasting (Prophet, ARIMA, regression models), contributing to a 12% revenue uplift in 2021 through optimized pricing and inventory. Automated recurring reporting with Python scripting and AWS Lambda, reducing manual workload by 40% and improving accuracy. Led data governance initiatives with AWS Lake Formation, implementing data quality checks, lineage tracking, and access controls to meet compliance standards. Integrated external datasets (Nielsen, social media sentiment, market data) with internal sources, enriching analytics for competitive intelligence and localized merchandising. Partnered with DevOps to deploy monitoring solutions via CloudWatch, improving reliability and proactive issue detection. Conducted churn and cohort analysis with SQL and Python, helping identify high-value customers and shaping retention strategies. Implemented CI/CD pipelines and version control with Git, AWS CodeCommit, and Jenkins for analytics and reporting workflows. Performed A/B testing, statistical significance testing, and confidence interval analysis, enabling data-backed marketing and product feature decisions. Built executive-level financial reporting dashboards for C-suite, tracking revenue, margins, ROI, and supporting quarterly board reviews. Acted as a liaison between business stakeholders and technical teams, ensuring analytical findings were translated into strategic business outcomes. Delivered training sessions for junior analysts on AWS tools, SQL, visualization best practices, fostering a data-driven culture. Client: Allstate, Northbrook, IL Sept 2015 - Dec 2018 Role: Data Analyst Responsibilities: Conducted data analysis on policyholder behavior, claims trends, and underwriting data, supporting insurance pricing and risk assessments. Leveraged AWS services (S3, Redshift, Athena, Glue) for large-scale insurance data processing in a secure cloud environment. Built and maintained dashboards in Tableau, Power BI, and SAS Visual Analytics, improving visibility into KPIs, claim volumes, and customer retention. Designed ETL pipelines using AWS Glue, SQL, and Python, automating ingestion from CRM systems and actuarial datasets. Collaborated with actuaries to develop predictive models in SAS and Python (pandas, scikit-learn) for churn prediction and fraud detection, reducing fraud by 18%. Utilized SAS statistical procedures (PROC REG, PROC LOGISTIC, PROC GLM) to support actuarial pricing models and dynamic customer segmentation. Optimized SQL queries in Redshift to improve reporting performance by 30% and deliver near real-time insights. Conducted cohort analysis and customer segmentation with SAS and SQL, directly improving cross-sell campaign effectiveness. Standardized structured & unstructured data integration across policy, claims, and customer interaction systems using AWS Lambda + Python. Participated in data governance initiatives, ensuring HIPAA and regulatory compliance with SAS data quality checks and metadata management. Applied statistical anomaly detection in SAS and SQL to flag unusual claim submissions. Partnered with DevOps for secure deployment of analytics workflows using AWS IAM, EC2, and CloudWatch. Delivered ad hoc SAS/SQL reports to support sales, marketing, and claims leadership in aligning business goals with analytics. Maintained data dictionaries & metadata repositories for transparency and cross-team usage. Regularly evaluated new BI and ML tools, helping Allstate evolve from SAS-heavy workflows to cloud-native Python/AWS analytics. Provided training on SAS, SQL, and Tableau for junior analysts, strengthening enterprise adoption of analytics practices. Client: Ooma Inc, Sunnyvale, CA Feb 2013 - Aug 2015 Role: Data Analyst Responsibilities: Partnered with engineering, operations, and marketing teams to analyze large-scale telecom datasets, improving operational efficiency by 20%+. Built and maintained dashboards and reports in Tableau and Power BI, providing executives with real-time visibility into call quality, churn, and customer KPIs. Executed end-to-end data analysis projects: data extraction (SQL Server, Oracle, MySQL), ETL transformations, and statistical evaluations. Conducted churn and customer behavior analysis using Python (pandas, NumPy) and R, leading to data-driven retention and loyalty strategies. Developed predictive models (logistic regression, decision trees) to forecast call drop rates and optimize network resource allocation, improving service quality. Implemented data validation and cleansing routines, improving reporting accuracy and consistency by 30%. Performed market segmentation analysis, uncovering key demographics and usage trends to support pricing and customer acquisition strategies. Supported telecom feature experimentation by designing A/B testing frameworks and analyzing product adoption outcomes. Conducted VoIP/call quality analytics, identifying network bottlenecks and recommending improvements that enhanced service reliability and customer satisfaction. Designed ETL pipelines for structured/unstructured data to standardize telecom analytics workflows. Assisted with database migration projects, validating data integrity and ensuring compliance with industry standards. Produced trend analyses for monthly/quarterly business reviews, highlighting growth opportunities and operational risks. Delivered training sessions to stakeholders on interpreting BI dashboards and reports, fostering a data-driven culture across the company. Maintained documentation of data models, processes, and business logic, ensuring scalability and knowledge transfer across teams. Keywords: cprogramm continuous integration continuous deployment quality analyst artificial intelligence machine learning business intelligence sthree active directory rlang trade national California Illinois Minnesota New York Tennessee