Jul 2021 – Present · 4 yrs 11 mos

Jun 2021 – Present · 5 yrs

Aug 2020 – Present · 5 yrs 10 mos

Dec 2018 – Feb 2021 · 2 yrs 2 mos · Gurugram, Haryana, India

• NLP Chatbots: (Insurance Domain – Marsh, USA)
• o Worked as a Data Scientist for Marsh Digital Labs (R&D) to build next-gen insurance product solutions.
• o Developed a generalized end-to-end chatbot solution by designing generic schema, building knowledge graph on Neo4j and training data using RASA to cater user queries by graph traversing.
• Enhancing capabilities like Context Handling, Coreference Resolution, Synonyms/Alias Recognition
• o Developed Client Company’s Internal Colleagues Bot “Marsha” to automate internal processes.
• User queries about info of other colleagues by name/designation/job role given in PDF format using RASA & SpaCy, achieved accuracy of around 97%. {well received by CEO of Marsh}
• User queries related to service desk like unlocking phone and info regarding company’s hackathon using AWS Lex & designing custom features using AWS Lambda
• o Automated Regression Testing of Chatbots:
• Automate the testing process by writing Python scripts to minimize validation testing time leading to improved accuracy by 25%.
• Predictive Modelling/Machine Learning:
• o Loss Projection Modelling: Predicting the total incurred loss (claim severity) for open and closed claims on 7 years of 2M historical data, features like nature of injury, cause of loss, body part of injury, claim description, claim closing date etc. Also, Classify claims based on level of complexity
• Actively involved in Sprint Planning, Creating User Stories, Grooming sessions, and daily Agile Scrum meetings. Used JIRA as agile project management tool to follow the software development project life cycle.
• Regular interaction with global clients (SMEs, Data Architects, Product managers) to understand business needs and functionalities for various project solutions.
• Conducted Data Science with Python 45 days classroom training course for the
• project team of 30+ corporate professionals of non-data science background.

Feb 2018 – Dec 2018 · 10 mos · Bengaluru Area, India

• NLP: (HR Domain)
• Worked on company’s main product HireAlchemy: Solution for Talent Acquisition & Workforce Optimization
• o Data Extraction: 20M resumes & 6M job descriptions from ElasticSearch of different clients.
• o Data Normalization: Normalized/Standardized Primary and Secondary Skills of candidates using NLP techniques such as rule-based algorithms and regular expressions.
• o Developed various scripts for feature extraction from unstructured data of resumes and jds.

Apr 2017 – Present · 9 yrs 2 mos

Jun 2016 – Feb 2018 · 1 yr 8 mos · Noida Area, India

• Predictive Modelling/Machine Learning: (HR Domain)
• o Candidate Approachability Problem: Build a prediction model with an accuracy of around 75% to predict candidate willingness to go for a job switch based on Business Development calls success.
• NLP: Chatbot POCs (Published Papers)
• o Developed QA bot by extracting intent and entities from user query using ontology & Neo4J Graph DB
• Data Analytics: (Published Paper)
• o Health Insurance Domain: Frequent Pattern Mining/ Fraud Detection to identify claims lifecycle anomalies/outliers and trends using statistical techniques.

Jun 2013 – Aug 2015 · 2 yrs 2 mos · Noida Area, India

• Data Analytics: Finance Domain
• o Performed the exploratory data analysis (univariate, bivariate, multivariate) for various features of the client’s dataset by plotting graphs and applying statistical modelling techniques using Python.
• o Worked on feature engineering, feature selection and feature extraction before data modelling.
• o Prepared power point presentations to demonstrate the data analysis to the clients.