Himanshu Goyal

Director of Engineering

India8 yrs 6 mos experience

AI ML PractitionerAI Enabled

Key Highlights

Led data science team at CARS24 for innovative solutions.
Built insurance risk models improving pricing strategies.
Created impactful machine learning applications across industries.

Stackforce AI infers this person is a Fintech Data Science Leader with expertise in machine learning and data-driven solutions.

Contact

himanshukumargoyal1273@gmail.com LinkedIn

Skills

Core Skills

Machine LearningData Analysis

Other Skills

AnalysisAnalytical SkillsAspen HYSYSAspen PlusBig DataBusiness AnalysisCC++Chatbot DevelopmentChemical EngineeringCloud ApplicationsCloud ComputingD3.jsData ModelingData Warehousing

About

As the Director of Data Science at CARS24, I lead a team of talented data scientists and engineers to deliver innovative solutions for the online used car marketplace. With a background in chemical engineering and business education, I have a unique perspective on applying data science to solve real-world problems and create value for customers and stakeholders. My passion is to use machine learning to extract actionable insights and shape impactful business strategies. At Acko, I built and validated insurance risk models that improved pricing and discounts for users and assets. I also used customer digital footprints to predict the buyout ratio and optimize the marketing campaigns. I have a track record of turning complex data into strategic advantages across various industries, including manufacturing, real estate, insurance, and fintech.

Experience

8 yrs 6 mos

Total Experience

1 yr 8 mos

Average Tenure

Current Experience

Cars24

Director Data Science

Apr 2023 – Feb 2024 · 10 mos · Bengaluru, Karnataka, India · On-site

Acko

2 roles

Lead Data Scientist

Promoted

Apr 2021 – Apr 2023 · 2 yrs

Data Scientist 3

Nov 2019 – Apr 2021 · 1 yr 5 mos

1. 𝐈𝐍𝐒𝐔𝐑𝐀𝐍𝐂𝐄 𝐑𝐈𝐒𝐊 𝐌𝐎𝐃𝐄𝐋𝐋𝐈𝐍𝐆:  Building improved insurance risk model based on user's demographics and Asset features. This risk model is used for determining insurance pricing and discounts.
Engineered meaningful features from claims data to assess the frequency and the severity of the events.
Built GLM-based and state of the art ML based insurance risk model to predict the frequency and severity for a combination of user and asset.
Validated through meaningful insurance metrics (e.g. lift charts, Double lift charts, Gini Index)
Developed the API endpoint to utilise the predictions for pricing calculations.
2. 𝐂𝐔𝐒𝐓𝐎𝐌𝐄𝐑 𝐁𝐔𝐘𝐎𝐔𝐓 𝐑𝐀𝐓𝐈𝐎:  Using customer digital footprints to predict the probability of conversion for a generated quote. This helps to re-prioritise the leads by Customer Support centre.
Applied unsupervised ML algorithms (Gaussian mixture modelling, hierarchical clustering etc.)  to draw meaningful insights from the features.
3. Built a reusable big data pipeline to query, clean, aggregate digital footprints data for more than 1 million users using PySpark.

Machine LearningData AnalysisPythonUnsupervised LearningFeature Engineering

Clearscore

Data Scientist 3

Nov 2018 – Nov 2019 · 1 yr · London Area, United Kingdom · Hybrid

Dhfl general insurance limited

Data Scientist 2

May 2017 – Nov 2018 · 1 yr 6 mos · Mumbai Area, India

As a Part of Digital Transformation Team,
Strategised and developed backend for Customer Website (www.dhflinsurance.com) and Insurance Agent's Portal using DJANGO framework in Python with a team of 4 developers.
Created in-house Chatbot for the assisting Customer Queries and linked it with Google Assistant for voice activation. It reduced the call centre traffic by 15%.
Worked on Car Telematics data analysis and created ML based Driver Scoring by extracting driving patterns from the recorded data. It helped in understanding the user-risk based on driving behaviour, thereby offering better pricing to low-risk customers.

DjangoMachine LearningData AnalysisChatbot Development

Hdfc red

Data Scientist

Aug 2015 – May 2017 · 1 yr 9 mos · Mumbai, Maharashtra, India

As a part of Data Science Team,
Created Chatbots to understand the requirements of the customers and suggest relevant real-estate properties. It reduced the lead-generation time.
Created a Recommendation Engine (patent pending) based on User Features and their digital footprints
Created a NLP & Text generation model based on RNN Architecture to autocomplete the search queries of the users

Chatbot DevelopmentRecommendation EngineNLPMachine LearningData Analysis