Austin, Texas, United States
Data is the information at an instance. Depending on the length of the instance, different types of data gets generated. The only reason Humankind was able to evolve so much wrt Technology is only because of Data that's getting generated every millisecond of the day. The importance of building data centric systems which scale have been monumental for our growth as well. The scale, size, types of data is only going to increase exponentially, so the skillset that's required to interpret it, to manipulate it, and to let the DATA talk is of great demand. With my prior experience, and understanding of Data and Analytics, I have a unique combination of experiences ranging from Using data for Business Intelligence and Driving decision making (at the exec level) to predicting demand, cost, incremental, & budget to make sure the businesses run smoothly, I have fulfilled various responsibilities across a span of 4 years in India. Use-cases ranging from Fraud detection, Audience group specific marketing & ads, Loyalty, CRM, E-commerce, Data driven Product Management, Data driven process optimization and many more across Pharmacy, Retail, E-commerce, and Post-Silicon Validation domains. My skillset ranges from SQL, Excel, Python, DBT, Airflow, Python ML libraries (sckit-learn, pytorch, etc...), Airflow, Distributed computing, Spark, Power BI, Tableau, Looker, Data Warehousing, and so many more. The only objective of this skillset is to enable the retrieval of insights, actions, recommendations from the data. The way that I pursue any problem is by using Divide and Conquer. I break down any complex problem into small distinct parts which together combine to solve the problem at hand. The endorsements in my LinkedIn account serve as a proof of truth for all the work I have established till now. With my Masters in Data Analytics from SJSU, I only look forward to learning, and upskilling further. I am always up for any interesting discussion, opportunity or simply a social event. I am available at [email protected], so please don't hesitate to reach out :)
1. Create Streamlined Executive Reporting for Manufacturing testing of AMD's DCGPU MY INSTINCT PRODUCTs ranging from MI3X, to the latest Open Rack Solution Helios MI4X 2. Identify High Impact Problems within Manufacturing at various supply chain levels to improve Quality and Reliability of the MI Products 3. Scripting and Automation to boost data driven decision making in Manufacturing Processes of the MI Boards using HBM level MFG info 4. Optimize Data Driven Customer RMA returns, to catch the issue during MFG at SLT, OAM, and UBB8 BOARD level testing 5. Day to Day Skills: Python, SQL, Power BI, Jump, VS Code, Snowflake, ETL/ELT, DBT
- Will be working on iPST - Intelligent Power State Transition tool, to optimize the existing manual processes for efficiency using Gen AI, Classical Data Mining and Deep Learning techniques, along with a Plethora of Statistics
- Created Training and Inference APIs banking on a Semi-Supervised Methodology to predict the status signals, thereby reducing debugging time by 25% - Created a Data Model to effectively model a SCANDUMP into status signals, Transaction Qs, FIFO Qs - Adapted Post-Si Specific Dimensionality reduction by utilizing the Physical SOC architecture, there by reducing the signals from 100M to < 10k - Collaborated with diverse SMEs, Validation Engineers, and other stakeholders to always know the customer inputs, to develop a completely scalable method which shifts left from the existing domain knowledge and dependency on Experience
Happy to assist Professor Eduardo Chan in Grading, Assisting the students for DATA 200: Python for Programming, and DATA 240: Data Mining for the fall semester Assisted with Grading for a class of 40 students for Python for programming, and Data Mining
After completing my Research Project on Urban Audio Classification with ESC-50 and UrbanSound8k datasets throughout the last semester, I was given the opportunity to take up the position of Instructional Student Assistant for DATA 270 under the guidance of Prof. Eduardo Chan.
Enabled automated SKU health monitoring for 300K+ SKU’s through Iceberg-based audits, PySpark Data transformations, visualized demand metrics on Power BI, helping operations leads respond to pricing discrepancies, stock dips, and forecast deviations in real-time ● Built supply chain intelligence pipelines using SQL, Redshift, PowerBI and Iceberg to forecast demand, detect pricing anomalies, and assess supplier lead time variability — reducing stockout rates by 15% and enhancing fulfillment accuracy across 2000+ Vendors ● Orchestrated multi-layered ETL workflows in Airflow, blending real-time stock movement with historical demand metrics in AWS Redshift, and visualized findings via Power BI dashboards for real-time operations, finance, and supply chain teams ● Designed 3+ data models for dynamic inventory and logistics simulations, enabling business teams to assess stockout probability and shipment slippage across geographies using Monte Carlo simulations and threshold-based logic in Python