Professional Contributions
Highlights
- Streamlined the trouble ticket process to eradicate low-confidence ML outputs, increasing the probability of identifying the root cause by 30% and decreasing manual effort by ~20%.
- Aligned standard operating procedures for feature identification of 4 machine learning models.
- Automated mail triggers using SharePoint and weekly reports sent to fulfillment centers, reducing manual effort by ~70%.
Responsibilities
- Developed a statistical model using Python to identify and categorize inventory count discrepancies caused by ML prediction errors, enabling targeted corrective actions and improved training at fulfillment centers.
- Analyzed inventory & FCSKU datasets queried using SQL, followed by data wrangling, identified the 'Scan while Stow' issue, and suggested preventive measures, leading to an ~8% decrease in low-confidence ML predictions.
- Led a team to develop Dislike, from definition to execution, defining metrics to track error rate and turnover time caused by quality issues, streamlined the process by bucketizing issues, setting up an SOP, and implementing product development based on data-driven insights.
- Analyzed gaps in the Dislike Mechanism using statistical modeling and identified root causes of high turnover time, presented logic changes for auto-generated corrective action emails to stakeholders, leading to a 30% improvement in turnover time and a ~20% reduction in manual effort.
- Identified root causes of incorrect image segmentation by Canvas, a vision-guided robot, implemented changes to improve product performance, enhancing its ability to detect movable objects and obstructions.
- Automated weekly reports using Microsoft Power BI, Python, and Excel, reducing the manual effort by ~70%.
- Led alignment of standard operating procedures (SOPs) across three machine learning models as the technical expert, mentoring and managing team members on feature identification aligned with product and business needs.
Project Journey
-
TRON
A model which is used to predict which bin a particular product is stowed into in a pod using videos at the Amazon Fulfillment Centre.
-
Proximity
A model which is used to identify human bodies in images and determine if the social distancing principles were followed at the Amazon Fulfillment Centres.
-
Robin - Damage Inquiry
A model used to classify whether a package in an image is damaged or not and the type of damage.
-
Photo Validation Process
A model used to check the compliance and safety of the pods and trays in the trucks using the images.
-
Package Zero
A model which checks whether a product is within or outside the package using images.
-
Canvas
A model used to identify movable objects in an image which is used by small size robots for moving directions.
-
SPA - FRB
A model used to classify whether a robotic arm is able to pick and place an item correctly using video based data.
Accreditations
- Completed Lean Six Sigma Yellow Belt Certification by ACES Academy
- Recognized as Champion IDS Research Analyst (Q2 2020)
- Recognized as Champion Auditor - Silver (Q3 2020)
- Recognized as Champion Auditor - Gold (Q1 2021)
- Recognized as Most Valuable Player (2020)
Key Solo Accomplishments
- Auto TT Mechanism: Proposed logic changes, reducing manual effort by ~7 hrs/week. Awarded Lean Six Sigma Yellow Belt Certification.
- Report Automation: Created automated dashboards for defect analysis using Microsoft Excel and macros.
- Feedback System: Developed workflows with SharePoint Designer to ensure timely acknowledgment and high program accuracy.
- SOP Changes: Simplified Dislike Standard Operating Procedure to enhance reporting of hardware problems at fulfillment centers.
Connect with Me!
© Lakshmi Keerthi Gopireddy