Kuntal Chakraborty

Collaborating on a private Agentic AI initiative focused on building intelligent, autonomous data engineering solutions using Large Language Models (LLMs) and AI agents. Designing and developing scalable data ingestion and transformation pipelines to support AI-driven workflows. Building Python-based utilities for data extraction, cleansing, validation, and metadata generation from structured and semi-structured data sources. Developing reusable ETL components to prepare enterprise data for Retrieval-Augmented Generation (RAG) and AI agent workflows. Working with Azure Databricks and PySpark to process large datasets and optimize data pipelines for AI applications. Integrating REST APIs and enterprise databases to provide AI agents with reliable, real-time business data. Implementing data quality checks, schema validation, and automated monitoring to ensure high-quality datasets for AI models. Managing source code using GitHub with version control, code reviews, and CI/CD best practices. Evaluating prompt engineering techniques and agent orchestration frameworks to improve automation and decision-making capabilities. Creating technical documentation, architecture diagrams, and workflow documentation using Miro and Visio. Collaborating with AI engineers and data professionals to prototype enterprise use cases involving intelligent document processing, data summarization, and workflow automation. Researching and experimenting with modern Agentic AI technologies, including multi-agent workflows, vector databases, semantic search, and Retrieval-Augmented Generation (RAG), to support enterprise data engineering solutions. Environment: Python 3.11, Azure Databricks, PySpark, Azure Data Lake, SQL, GitHub, REST APIs, VS Code, Jupyter Notebook, OpenAI APIs/LLMs, LangChain, LangGraph, Vector Database, RAG, Miro, Visio, CI/CD.

• Supported a hybrid integration architecture using IBM ACE to orchestrate data workflows across DB2, Oracle Cloud, Azure Data Lake, IWMS platforms, and Dynamics 365. • Designed and implemented Spark-based ETL/ELT pipelines in Databricks using PySpark to process large-scale financial and mortgage datasets. • Developed executive-level Power BI reports using advanced DAX measures and M-Code transformations for debt analysis across 2500+ locations. • Built and optimized Gold Layer datasets to support Regulatory Reporting, Actuarial Analysis, Financial Balancing • Built predictive models in Databricks (PySpark) using XGBoost, scikit-learn, and Azure ML SDK to forecast real estate pricing trends and housing demand. • Ensured schema governance and metadata management aligned with Unity Catalog standards. • Monitored job execution, optimized cluster performance, and reduced pipeline latency by tuning Spark configurations and partitioning strategies. • Employed Azure Machine Learning to build, train, and deploy predictive models for real estate market forecasting, price trends, and demand estimation. • Applied advanced data wrangling: normalization, deduplication, missing value imputation, feature engineering, and unstructured-to-structured transformations for IWMS and mortgage datasets. • Created detailed process mapping diagrams (Miro/Visio) for end-to-end data pipelines. • Developed CI/CD workflows with GitHub to maintain real-time data synchronization across on-prem, cloud, and IWMS systems. • Integrated Oracle data sources with SQL Server and cloud platforms across hybrid multi-cloud environments. • Partnered with Product Owners, Data SMEs, and reporting teams to translate complex business logic into scalable Spark solutions. Environment: Azure Databricks, PySpark 3.4, Python 3.11, Delta Lake, Azure Data Lake, Snowflake, SQL Server, Oracle Cloud, GitHub, Power BI, Azure ML, Azure Cognitive Services, IBM ACE, IBM CP4I, IWMS, IBM SPSS, CI/CD, Jira/Confluence and Miro

Project : Data Transformation and Intégration – Azure DB to Cloud Architecture Responsibilities: • Conducted Gap Analysis for 23 wind turbine sites missing data from IP21 Historian into Delta Lake, calculating TAG counts and building gap sessions in Agile time clusters. • Backfilled historical data for wind turbine sites in Calgary and Sarnia, ensuring data accuracy and continuity. • Developed PySpark 3.2 notebooks (Python 3.6) and SSIS ETL packages for Time Weighted Average (TWA) computations, improving KPI delivery through Power BI dashboards. • Designed Azure Analysis Services models and Databricks-based Python transformation scripts for unified reporting. • Managed data workflows using Apache Airflow, integrating TWA tasks to visualize pipeline dependencies, execution logs, and completion metrics (Team Lead). • Validated and migrated data from legacy SQL tables to Azure Delta Lake, creating intermediate and temporary staging layers. • Utilized PySpark Streaming and SSIS for real-time and batch data integration, supporting near real-time analytics and decision-making. • Designed Miro flow diagrams and architecture blueprints for Wind, Gas, Thermo, Solar, and Heat power generation units across all fleets. Environment: MS SharePoint, PuTTY, GENOS inbuilt system, MIT, TSQL, SQL Management Studio, GIT Hub, Bit Bucket, PySpark, Azure DB, MIRO BOARD, AZURE Platform and Apache

Project : # Marketing - Digital and Model Score (MDEC) - POD2 and POD4, Cash Project (MDP) – POD1 Responsibilities: • Developed code according to template designing data frame using PyCharm IDE tool and Jupiter Lab for PySpark coding. • Validated data from AKORA HIVE tables as source and developed physical intermediate and temporary tables for Azure platform using SQL queries in ADB (Azure Data Bricks). • Unit Testing (white box), Data Mining, Data Defining, Data Profiling is part of day-to-day activities. • Created, initiated, and updated confluence page and designed MSVISIO flow diagrams and charts for Cash Flow project. • Code conversion - to convert SAS code DB2 to AKORA HUE HIVE big data for Digital and Model Score jobs to get exact scoring values for each customer. • Stored Data in our BIT BUKCET private repository after code review has been completed and committed. Environment: PyCharm, PuTTY, Control-M scheduler, Batch Scheduling, WinSCP, SAS Guide, GIT Hub, Data Bricks, Bitbucket, PySpark, Jupiter Lab, JIRA and CONFLUENCE, AZURE Platform and Apache Spark

Project : # Retail-RRMDM Digital Crystal Jewell Project Responsibilities: • Created Stored Procedure to drop objects unnecessary tables or procedures if exists in databases OOS in Veracode scanning report. • Integrating data changes in DEV Servers with Visual Studio SSDT packages to harmonize backend server with front end applications to maintain data integrity in production server. • Created CR for all drop objects and migrated design into cloud Azure Platform from legacy to current updated stored procedure. • Developed Reusable Assets in Jenkins for continuous Integration and Automation Build Verification • Provided knowledge transfer to the team offshore and onshore team members. • Reviewed the test plan and test results, created and tracked the defects JIRA • Performed System Testing, System Integration, Component/Unit, UAT/ end to end system testing • Re-design all CAP procedures and Material Exception Retail Risk Management Data Mart (RRMDM) Veracode High Risk mitigate by design or migrate by design. • Created ETL packages with different data sources (SQL Server, Flat Files, Excel source files, XML files) and then loaded the data into destination tables by performing different kinds of transformations (snowflake) using SSIS packages. • Performed analytical activities in reporting using Tableau - Dashboard presentation to showcase KPI including graphs and charts focusing on Retail Risk Management Data Mart maintaining SLA. • Experienced with SQL Injection to protect database from outside vulnerabilities using Veracode Scanning Tool using SAST, DAST and Manual method protecting TD RRMDM data. Environment: SSMS 2016, Control-M scheduler, Batch Files Scheduling, Veracode, SSDT 2016, Tableau, (Azure Data Bricks), SSDT Visual JIRA and CONFLUENCE Studio 2016.

Responsibilities: • Created workspace for the client to get provision to integrate Data Warehouse with Data Lake which in return save investment and increase revenue 5M this budget fiscal year. • Ingested data and Explore data coming from 85+ connector from different sources in forms of Flat Files, SAAS Files, Cloud files, TSQL non structured files, R files etc. and created pipelines through Azure Analytics tool. • Created test plans and UAT testing functionalities for modules like Deposit, Payment / Invoice, Tax and Government Fee • Create and maintain performance testing initiatives and adhering to best practices. • Created ETL packages with different data sources (SQL Server, Flat Files, Excel source files, XML files) and then loaded the data into destination tables by performing different kinds of transformations using SSIS packages. • Experienced with Copy data bringing Big Data, Data Warehousing, and Data Integration from all different geographical locations into single service E2E analytics at cloud scale. Environment: SQLServer2017 Data Warehouse, AWS Code Build, Azure DevOps, Power BI and Spark, Atlassian JIRA, and CONFLUENCE

Responsibilities:  Designed SSIS packages to Extract Data from different sources like Flat Files, Excel Files, SQL Server and Snowflake Transformation to load into Dimension and Fact tables in Data Warehouse using SSIS.  Implemented SSAS Snowflake Cubes, Facts and Dimension tables to perform analysis services.  Involved in Migration of Databases from SQL Server 2008 to SQL Server 2012 using SSIS.  Involved in Data Warehouse Migration from different Sources to Destination like Extract-Clean-Standardize- Enrich-De Dupe-Transformed-Upload/Import  Designed and implemented Data Marts, Facts, Dimensions, OLAP, OLTP cubes using dimensional modeling standards in SQL server.  Involve in Swisscom CES project to support NetProvision MPM to SRE Flow. * Deliver Bell Canada NetProvision custom components and IPACT Batch Mode Activation.  Used Bulking techniques, Table partitioning to increase database performance.  Understood the design of star schema and snowflake schema.  Used various data flow transformations like conditional split, look up, derived column etc.  Generated various reports with Drilldowns, Drill-through, calculated members, dropdowns from the cubes by connecting to Analysis server from SSRS.  Use ODBC tools transfer data Hive to SSIS. Environment: SQL Server 2012, SSIS, SSRS, SSAS, T-SQL, Atlassian JIRA Bamboo, MS Office, TABLEAU

Responsibilities:  Cleansed and transferred data from various OLTP data sources, such as Oracle, MS Excel, Flat Files, CSV files into SQL Server 2014 by using SSIS packages.  Designed and implemented ETL system (suite of batch jobs) to process, report, and warehouse transactions for AML/ATF compliance  Created ETL packages with different data sources (SQL Server, Flat Files, Excel source files, JSON and XML files) and then loaded the data into destination tables by performing different kinds of transformations using SSIS packages.  Transmitted data in web applications (e.g., sending some data from the server to the client, so it can be displayed on a web page, or vice versa).  Created Roles for SSAS Tabular Model provide additional Security at the row-level by using row-level filters and designed perspectives and hierarchies.  Created complex SSAS Snowflake cubes with multiple fact measures groups, and multiple dimension hierarchies based on the OLAP reporting needs.  Created and maintained logins, roles and permissions on the SQL instances and databases and provided required privileges according to the user requirements.  Involved in development, deployment, scheduling, and troubleshooting SSIS packages.  Involved in complex SSIS package development with Microsoft BI standards and Development strategies to pull data from various data sources such as (SQL Server, Excel, and Flat file).  Closely monitored Data Hub and Info Center team and helping them for daily reporting  Knowledge transferred to performance, automation and TDM team through training sessions and walk- through.  Always tried to give 100% effort to reduce compliance in SLA/KPIs in SOW. 2-3% deviation sometimes occurs with respect to SOW. Environment : Guide Wire 9.0, DB visualizer, SQL Server Management studio, SQL Server Business Intelligent studio (BIDS), SOAP UI, Oracle Data Hub, Data Bricks, Info center, Atlassian Bamboo, JIRA and CONFLUENCE

Responsibilities:  Extensively worked with SSIS tool suite, designed, and created mappings using various SSIS transformations like OLEDB Command, Conditional Split, Lookup, Aggregator and Multicast  Used SSIS to create ETL packages to validate, extract, transform and load data to data warehouse.  Used Joins, correlated and non-correlated sub-queries for complex business queries involving.  Developed complex Stored Procedures and views to generate various Drill-through reports, parameterized reports, Tabular reports, Matrix reports and linked reports using SSRS.  Created clustered and non-clustered indexes on tables for new database to faster retrieval of data for the reports.  Used TVD_EMC logs tool to get CTN and BAN for successful order submission during data creation.  Used CRM Maestro in creation of consolidated account for customer having bundle package HSI, IPTV, Cable, STB and Home Phone  Used SOAP UI, ERS and QA data tool to get SIM, IMSI numbers allocation for CSM V21 test data creation for wireless customer used for projects.  Worked in billing project creating reports for Business Regular customer, who can be able to view their invoices at Browser and App both in HTML and PDF formats.  Worked extended hours throughout the weekdays and in also in weekends – whenever the project team required.  Majority times involved in device unit testing in different Browsers and several build versions in APPs with various device models of APPLE, ANDROID, MacBook, and TABLET Environment: SQL Server 2008R2, SSRS, SSIS, SSAS, Crystal Reports, MS Excel, JIRA, HP-ALM 12.21, RQM, RCQ, Jenkins, Outlook Express, AMDOCs – CRM Maestro, CSM – V21, ERS and TVD_EMC logs tools, SOAP UI, Yammer, and Microsoft SharePoint.

Responsibilities:  Performed backup/restore, database objects such as tables, procedures, triggers, constraints, indexes and views, user security management.  Responsible for developing processes, automation of maintenance jobs, tuning SQL Server, locks, and indexes configurations, administering SQL Server security, SQL Server automatic e-mail notification and SQL Server backup strategy and automation.  Written complex queries and stored procedures to fulfill the enhancement requirements of data base migration project.  Performance tuned some of SQL queries and stored procedures using SQL server Profiler.  Designed and developed different type of reports including Tabular forms, Matrix (cross Tab report) form, Conditional, List items, Charts and Sub reports, Drill down reports.  Defined parameters for parameterized reports and implemented cascading of the parameters.  Involved in project planning also schedule for database module with project managers.  Managed SQL user logins, database user accounts, server roles, database roles and its permissions.  Improved Performance by creating Clustered and Non-Clustered Indexes and by optimizing the T-SQL statements using SQL profiler.  Designed and developed different type of reports including Tabular forms, Matrix (cross Tab report) form, Conditional, List items, Charts and Sub reports, Drill down reports.  Designed and implemented the warehouse data model, dimension, fact, and summary tables.  REST API is a way for two computer systems to communicate over HTTP in a similar way to web browsers and servers. Environment: MS SQL Server 2014, SSIS, SSRS, ORACLE, MS Office, TFS.

Projects:  SharePoint2013 - CIBC Today homepage, Release 2a and 2b, Mobile IID and United Way Fall Campaign 2014 & 2015 Responsibilities:  Created Test Cases for All Sites, Profiles, Communities, Search and Team Sites following project backlog.  Planned Test Strategies, Gap Analysis and helped in writing Test Plans and exit report for sign offs and attained GATING meet to represent our QA Testing Team  Worked closely with the Business Analyst, Developers and Test Lead to understand Business and Functional specifications and discussing defects in Agile architecture.  Executed Test cases both in English and French language with Admin and Non-Admin user Test Accounts using IE8.0 and IE 10.0 browser in dedicated VMs for QA testing.  Used HP ALM 11.52 Auto Generate Report features to track Daily Defect Status Report for Defect tracking.  Performed regression testing using QTP by executing test scripts and analyzed the results.  Submitted daily and weekly status report to QA Lead and Development Team.  Used Quality Center HP ALM 11.52 for Test management and for Defect tracking purpose, identified, tracked, and reported defects to the developers with proper priorities and severities.  Attained daily SCRUM meeting and followed by SPRINT bi-weekly basis.  Followed instructions as per project Dashboard and Bulletin Chart released daily basis.  Scripted Test Cases in HP-ALM in separate Iteration cycle in Incremental methodology and created automation for regression testing to run batch processing daily and weekly.  Planned Test Strategies, Gap Analysis and helped in writing Test Plans and exit report for sign offs.  Record breaking donation amount cross the upper limit tag $14000 crossed $16000 mark. Environment: HP-QTP, Outlook Express, F12Developer tools, Office Communicator and Microsoft Office, Snippet Tool, Selenium 3

Responsibilities  Worked closely with the Business Analyst, Developers and Test Lead to understand Business and Functional specifications.  Performed Usability testing, Sanity, Regression, and Functional testing during testing phase.  Executed Test cases, identified test data and created test data for testing Environment using Decision Table Technique  Used Quality Center for Test management and for Defect tracking purpose, identified, tracked, and reported defects to Developers with proper priorities and severities.  Provided effective and timely communication to all internal stakeholders relating to project concerns issues and quality.  Worked in the modules to test Site Front End, Application, Site Cash and Termination  Worked in sub-modules in Site Front End that were Hotspot Toolbar, ALV Grid Hotspots, General search, and Vacancy search.  Worked in sub-modules in Application (contract offer) that were to create application, add and change business partner, maintain business partner, reminders, building services, incentives, override, payment, change start date and change rental objects.  Worked in sub-modules in Termination that were Termination of a Suite, Termination cancels, Move-out, Termination of parking or locker.  Executed checkpoints, synchronization, regular expression, and data driven framework during automation testing.  Submitted daily and weekly status reports to QA Lead and Development Team Environments: HP Quality Center, HP-QTP, SQL Server, SAP R/3 BI, SAP Basis

Responsibilities  Reviewed and understood the requirements and design by organizing multiple review meetings with SME’s and Business Analysts  Reported document gaps to appropriate authors and requested for an updated version of the document.  Performed Sanity testing on the online application to ensure the stability of the build in test environment.  Executed test cases to validate Functionality of the online banking application by verifying functionality using both positive and negative testing approach: bill payments, cancel pending payments, transfer, email money, cancel pending email, view statements & transactions and get quotes.  Performed extensive penetration and security testing with the help of network and Developers on the banking application.  Performed Usability testing to verify interface, navigation, and intuitiveness of application. Environments: Quality Center, Quick Test Professional, XML, HTML, DHTML, JavaScript, SQL Server, MS Access, Microsoft SharePoint, and Windows XP