Somesh Manda

Developed and delivered 13+ data products like Batch Genealogy, RIMA, Batch failure prediction in Quality and Supply chain using Azure products.

Responsibilities:

• Managed OPEX of 1.5M USD and CAPEX of 6.5M-8M USD annually to drive business value thru data products for vaccines manufacturing and quality production
• Designed and delivered GXP compliant Cloud (Azure) platforms for enterprise data and Analytics in six months with two platform engineers
• Envisioned end to end automation of product quality reports across sites and products
• This includes end to end data pipeline with ML/NLP techniques
• This reduced cycle time of report generation from 4 months to 2 weeks for report, and majority of 2 weeks is just manual review
• Analyzed data and report proliferation across vaccines enterprise and recognized ~4500 person days of manual effort and 40+ tools to generate insight
• Developed business case to transform ways of working and delivered operational savings incrementally thru self-service BI and trusted reports automation
• DARA
• Directed enterprise vision to improve financial and operational efficiencies by leading a team of 75 engineers with varied skill sets like platform engineers, data engineers, BI engineers, Scrum masters, BI, testers, UI and ML engineers
• Reimagined manufacturing audits thru data driven workflows
• Experimented and designed risk assessment models by collaborating with Data scientists to advise site staff on potential audit risks
• Partnered with auditors to establish new ideas continuous audit and risk assessment by exploiting AI and data engineering techniques
• Designed automation for data ingestion and consumption of 736 equipment that are connected to enterprise network across 13 manufacturing sites
• Ingested 1000+ new data sets in less than 2 months and industrialized automation and daily refresh as a fully GXP qualified process
• Implemented first ever cloud based GXP qualified property graph (manufacturing BG) solution in industry using Azure Cosmos DB and Gremlin API
• Architected data pipeline that provides daily prediction of batch failures which has multiple million-dollar financial savings
• Automated finance data reporting for CFO tracking of Business performance across headcount, cost
• Recognized for diversity and inclusion mindset in working with teams that are spread across geographies, religions and cultures
• Provided oversight guidance different solution partners that include TCS, Wipro, Accenture, TechM, CTS and Smarttronix
• Received prestige’s CEO (IPTc) Award from executive leadership.

Led team of 20 data and cloud engineers to build data warehouse on AWS using RedShift. Established data warehouse standard and procedures.

Worked for Northern Trust to build enterprise data warehouse using Hadoop, Scala, Sqoop and other big data technologies.

Migration project from information to DataStage.

Worked with AWS EC2 and home grown clusters

Large amount of high volume Data from bidders and other networks was streamed daily to S3 Amazon platform from S3 platform, Data was loaded into Hadoop Cluster. This Data was maintained for 60 days for analysis purposes600-700 GB of Data was loaded daily into Cluster.

Aggregate Spend is the process used to aggregate and monitor total amount spent by healthcare manufacturers on individual Healthcare Professionals and Organizations (HCP/O) through payments, gifts, travel and other means

Also often referred to as Physician Spend Sunshine Law. Aggregate Spend” is total, collective, cumulative amount spent by healthcare manufacturers on individual Healthcare Professionals and Organizations (HCP/O) through payments, gifts, travel and other means.

Team Lead for various projects including Advertising, War Room and other projects. Projects include in migrating from Business Objects Data Integrator to Informatica.

ETL Team Lead for enterprise data warehouse containing all data pertaining to Deposits, CDs, Mutual Funds, Customer data Crucial, centralized reporting supporting Direct Marketing analysis, financial reporting, and regulatory compliance for $3 billion Federal Savings Bank.

Customer relationship management is business strategy to help enterprise to understand needs of its customers. As fundamental component of this strategy, Merck Standard Customer Repository (MSCR) is built to provide central repository for valid customer information for MERCK. MSCR also stores Alternate ID, address, license, credentials, corporate designation, and association’s information along with customer information. Data is populated into MSCR by external and internal loads. External data comes from various sources like AMA, OMS, AOA, IMS and CIMS.

Project is to build Sales data warehouse. Objective is to extract data stored in different databases such as DB2, SQL Server, Sybase and load finally into single data warehouse repository, which is in Oracle database.