Case study
Deloitte: machine learning for Fortune 100 clients
ML and data work on Fortune 100 engagements: an ML contact-center product for Pfizer that cut average queue time from 75 to 15 minutes, and a clustering model for GM that saved $17.5M.
Confidentiality. Details are under NDA; this describes my role and approach, not proprietary specifics.
Context
From 2020 to 2022 I was a Machine Learning Consultant at Deloitte, working on engagements for Fortune 100 clients including Pfizer, General Motors and Walmart. At the same time I took graduate coursework in data science through Georgia Tech's OMSA program. I was promoted to senior in my first year, in the top 5% of my new-hire class.
The problem
Each engagement was a different problem. During COVID-19, a contact center was overwhelmed and wait times were long. A manufacturer was carrying configurations that cost more than they returned. A retailer needed its data moved onto a new platform without losing the ability to run analytics at scale.
My role
I worked on the machine learning and data side of consulting engagements, including hands-on ML engineering: managing the development of ML products, designing and deploying models, and designing data architecture. I also synthesized market research, competitive analysis and operational data into recommendations clients could act on.
What I built / did
- Pfizer: managed the development of an ML-integrated contact-center product during COVID-19 that automated response generation for incoming requests.
- General Motors: designed and deployed a clustering algorithm (unsupervised machine learning) to identify and prune low-value configurations.
- Walmart: led a data migration to SAP S/4HANA and designed the data architecture for scalable analytics operations.
Results
- Pfizer: average queue time fell from 75 minutes to 15.
- General Motors: the configuration pruning saved $17.5M.
- Promoted to senior in year one, top 5% of the new-hire class.