Freelance data scientist · All case studies

Complaints that route themselves.

An NLP pipeline that reads inbound tickets, classifies the complaint, and sends it to the right queue — text representations plus classifiers, not a intern with a keyword list.

Client: Customer operations. Built by Dilshad Raza.

Inbound complaints and support tickets arrived as free text. Billing, outages, account access, and product defects landed in the same pile. Manual tagging lagged the inbox, and a regex for “refund” was not a taxonomy.

I built an automated classification pipeline: NLP text representations on the inbound copy, then machine-learning classifiers trained to parse the complaint and route it. The desk still sees the ticket. They no longer have to invent the category from a blank form.

Hire Dilshad Raza for similar freelance data science, machine learning, and AI automation in the UK, United States, and Australia.