Home

Srikar Duriseti - Data Engineer / Supply Chain Analyst

[email protected]

Location: Placentia, California, USA

Relocation: Anywhere in usa

Visa: F1-STEM OPT

Resume file: Srikar Duriseti_1775503223058.pdf Please check the file(s) for viruses. Files are checked manually and then made available for download.

SUMMARY Data Analytics Engineer with 5+ years of experience building modern lakehouse platforms, forecasting systems, and AI-driven analytics solutions across healthcare and enterprise supply chain domains. Skilled in Python, SQL, PySpark, and Microsoft Fabric for developing scalable pipelines, time-series forecasting models, and semantic analytics layers. Hands-on experience delivering production-grade Azure solutions including Databricks workflows, MLflow experimentation, and OpenAI-powered applications. Proven collaborator experienced in translating business requirements into reliable, data-driven platforms that improve operational efficiency and decision-making. SKILLS Programming & Querying: Python, SQL, PySpark, Spark SQL, Pandas, NumPy Data Engineering & Processing: ETL/ELT pipelines, Delta Lake, ADLS Gen2, Data flow Gen2, medallion architecture, semantic model, audit logging, fine tuning, incremental processing Cloud & Data Platforms: Microsoft Fabric, Azure Databricks, Azure Synapse, Azure Data Factory, Azure Data Lake, Cosmos DB, Azure SQL ML & Advanced Analytics: scikit-learn, XGBoost, MLflow, Prophet, time-series modeling, model evaluation (AUC, recall), feature engineering, fine-tuning Forecasting & Supply-Chain Analytics: Demand forecasting, inventory optimization, SKU-level forecasting, lead-time analysis, clustering for allocation Data Visualization & BI: Power BI, semantic models, dashboard development, KPI reporting, drill-downs, data storytelling DevOps, Streaming & Governance: Streamlit, Git, Azure DevOps, Airflow, Event Hub, Stream Analytics, Application Insights, Log Analytics, RBAC, Key Vault, data governance, HIPAA awareness EXPERIENCE Data Analytics Engineer | Quadrant Technologies, USA May 2025 Present Built a Microsoft Fabric medallion architecture (Bronze Silver Gold) integrating MySQL and CSV sources to standardize ingestion and reduce data preparation time for forecasting and optimization workflows by 45%. Developed incremental PySpark pipelines with schema validation, deduplication, and feature-ready transformations to prepare service call and technician datasets for predictive modeling and downstream analytics consumption. Designed Fabric notebooks to engineer features for service demand forecasting, enabling region- and technician-level predictions that supported proactive scheduling and resource planning. Built incremental PySpark MERGE pipelines with 24-hour lookback, schema validation, JSON flattening, and audit logging, ensuring clean, deduplicated data across all layers. Implemented time-series forecasting models to predict service call volumes, helping operations teams anticipate demand patterns and improve workforce allocation across service territories. Created inventory forecasting datasets using historical parts usage and service data, enabling predictive insights that supported supply planning and reduced repeat service visits. Delivered Power BI dashboards visualizing historical trends, demand forecasts, and operational KPIs, enabling stakeholders to monitor forecast accuracy and optimize technician workloads. Operationalized forecasting outputs into Fabric semantic models and governed pipelines, ensuring reliable refresh cycles and scalable analytics for supply chain and service operations teams. Worked closely with client stakeholders to translate predictive analytics use cases into production-ready Fabric solutions and supported handoff, documentation, and post-deployment adoption. Supply Chain Data Analytics Engineer | UnitedHealth Group, USA Jul 2024 Apr 2025 Analyzed procurement and inventory datasets across Optum supply networks using Python and SQL to identify utilization patterns, enabling cost optimization and improving supply availability across clinical and operational workflows. Developed demand forecasting models using PySpark and time-series techniques to predict medical supply consumption, supporting proactive replenishment strategies and reducing stockout risks across healthcare delivery centers. Built scalable ETL pipelines integrating supplier, purchasing, and utilization data into Azure-based analytics layers, enabling unified visibility into supply chain performance and improving reporting accuracy for operations stakeholders. Designed Power BI dashboards highlighting supplier performance, inventory turnover, and demand variability, enabling procurement teams to track KPIs and improve sourcing decisions across enterprise supply chain operations. Optimized supply planning workflows by applying clustering and segmentation methods to categorize facilities by utilization patterns, enabling data-driven allocation strategies and improving distribution efficiency across regional healthcare networks. Collaborated with sourcing, finance, and clinical operations teams to translate supply chain analytics into actionable insights, supporting data-driven procurement strategies aligned with UnitedHealth Group s operational efficiency initiatives. Data Analyst | CitiusTech, India May 2019 Aug 2022 Analyzed multi-source clinical and claims datasets using SQL and Python to support healthcare AI model benchmarking, improving use-case mapping accuracy and enabling faster identification of high-value predictive analytics opportunities. Structured normalized data models in Snowflake and relational warehouses to unify EHR, lab, and patient demographic feeds, reducing data preparation time for downstream analytics workflows by 30%. Constructed reusable ETL pipelines with Python and Spark to standardize ingestion of curated healthcare datasets, ensuring model-ready inputs and minimizing manual preprocessing across Medictiv evaluation cycles. Produced Power BI dashboards highlighting model accuracy, recall, and cohort segmentation, enabling stakeholders to prioritize disease prediction use cases and accelerate analytics-driven roadmap decisions. Validated dataset quality through automated profiling and rule-based checks, reducing inconsistencies in model training inputs and improving benchmarking reliability across multiple healthcare AI experiments. Partnered with data scientists and clinical SMEs to translate use cases like readmission prediction and chronic disease monitoring into analytical datasets, strengthening alignment between platform outputs and provider decision workflows. Enhanced metadata indexing using Elasticsearch and structured tagging, improving discoverability of healthcare AI models within the Medictiv directory and reducing internal search time for analytics teams by 25%. Authored comprehensive data dictionaries and governance guidelines aligned with HIPAA standards, enabling scalable onboarding for global analytics teams and improving data transparency across healthcare innovation initiatives. EDUCATION Master of Science in Data Science University of North Texas, TX May 2024 B.Tech in Aerospace Engineering Amrita University, India April 2020 PROJECTS Customer Churn Prediction Pipeline (PySpark, Databricks) Developed a churn intelligence pipeline using PySpark, MLflow, and feature versioning to generate predictive retention signals and deliver actionable insights through BI dashboards. Evaluated multiple classification models using AUC and recall metrics to select the optimal churn predictor, improving targeting precision and reducing customer outreach inefficiency. Real-Time Customer Intelligence Pipeline Built a streaming analytics platform using Azure Event Hub and Stream Analytics to enable real-time behavioral insights for adaptive marketing strategies. Integrated processed events into a low-latency Azure SQL layer with DirectQuery, enabling business users to visualize campaign performance with under 5-second delay in Power BI dashboards. Keywords: artificial intelligence machine learning business intelligence database Texas