Aug 2022 – Jul 2024 · 1 yr 11 mos

Jul 2021 – Aug 2022 · 1 yr 1 mo

Jul 2019 – Jul 2021 · 2 yrs

• Improved and curated Location Insights feature, which predicted and analysed insights and sentiment in location based data, trained models using DistilBERT and spaCy algorithms and coded framework using python3. Ensured smooth processing for data of 70+ clients spread over 10 verticals using kafka pipelines
• Developed Smart Intents which is an automated pipeline for training and deploying bots by collecting data from database, train using ML algorithm based on data size and then deploying the service on kubernetes dashboard
• Developed automated chatbot server integrating open source Rasa for production, which trained, inferred, updated and synced models across all instances.
• Implemented Louvain clustering algorithm based on similarity to be later used for data cluster recommendation for training models to detect intent and entity in a given user query using similarity index and state of the art sentence embedding.
• Developed article suggestion module with step-wise probing functionality using Faiss similarity index and MUSE to suggest articles with the description closest to the problem mentioned

May 2018 – Jul 2018 · 2 mos · Gurgaon, India

• Designed prediction models for Baseline profts and Incremental profts utilizing spend and membership behaviour
• Improved Retention program by implementing prioritization of willing-to-attrite customers based on Incremental Profts
• Analysed features based on intuition by studying correlation and ranks and plots for legitimacy using python
• Simulated the proposed prioritization on past one year's data showing positive impact in all revenue ranges