Virtual Assistant for Hiring Last-Mile Workforce
Submitted by Piyush Makhija (@piyushmakhija) on Saturday, 29 September 2018
Logistics companies, both old and new, have invested heavily in building an efficient frontline workforce to provide swift and convinient services to their users. Timely delivery is often a critical deciding factor for the ever-impatient customers to choose service A over service B. Hence, operations/logistic team is the key enabler here.
The attrition rate in large frontline teams is high, close to 75 percent annually. Yet most companies have aggressive growth targets, necessitating recruitment of high volumes of workers constantly. High-growth companies in this domain like Zomato and Swiggy, which are trying to grow by 50-60 percent by the end of 2018, need to recruit tens of thousands of delivery boys every month.
At Vahan, we have developed an AI-driven virtual assistant that helps logistics companies scale and automate their hiring process by leveraging the common addiction of messaging applications like WhatsApp and FB messenger.
In this talk, I will cover in detail how we developed a complete data collection and natural language processing pipeline for indian languages and built a chatbot over Whatsapp which is currently connecting companies like Dunzo, Dominos & Ecom Express with potential frontline workers and fulfiling the hiring requirements of this industry in a scalable and autonomous fashion.
- Introduction to the High volume Recruitment problem
- Breakdown of the problem in terms of technical challenges
- Natural Language Processing Pipeline
- Building a data collection pipeline
- Handling Variability in Data
- Understanding User Needs
- Edge Cases
- Performance and Evaluation
- The Road Ahead
Basic Knowledge of Machine Learning and Natural Language Processing is required to understand the contents of the talk
Piyush is a graduate from Georgia Institute of Techonology and is currently working as an NLP Engineer at vahan.ai. After passing out from IITR with a Bachelors in ECE, he started out his career as a 4G protocol engineer but soon got attracted towards the fast growing ML/AI domain. Over time he switched over to this domain and, after some exploration, found his interest in working with vernacular languages.
When he is not at work, he spends his time focussing on fitness and honing his skills with the guitar.