Rahul D.

AI Researcher

Bengaluru, Karnataka, India3 yrs 8 mos experience

Key Highlights

  • Expert in contextual bandits for ad optimization.
  • Proven track record in user response modeling.
  • Strong foundation in machine learning and data science.
Stackforce AI infers this person is a Data Scientist specializing in AdTech with a strong focus on machine learning.

Contact

Skills

Core Skills

Machine LearningData Science

Other Skills

AdAsset SelectionAlgorithmsApplied StatisticsAutomationBERT (Language Model)Bayesian StatisticsBayesian statisticsBusiness MetricsC++Click PredictionCommunicationComputer VisionContextual BanditsData AnalysisData Structures

About

Our team at Microsoft has been leveraging my expertise in contextual bandits and other ml models to enhance user response modeling within Microsoft Ads. With a Bachelor of Technology Electronics and Communications Engineering from IIT Guwahati, my role involves optimizing AdAsset selection and click prediction.

Experience

Meesho

Data Scientist 2

Aug 2024Present · 1 yr 7 mos · Bangalore Urban, Karnataka, India · Hybrid

Microsoft

3 roles

Data & Applied Scientist (L 60)

Sep 2023Aug 2024 · 11 mos · Bengaluru, Karnataka, India · Hybrid

Data & Applied Scientist

Jun 2022Aug 2023 · 1 yr 2 mos · Bengaluru, Karnataka, India · Hybrid

  • Microsoft Ads - User Response Modelling Team.
  • Working on AdAsset Selection, Click Prediction.
Contextual BanditsAdAsset SelectionClick PredictionMachine LearningData Science

Data and Applied Scientist Intern

May 2021Jul 2021 · 2 mos · Bangalore Urban, Karnataka, India

  • My internship was in the Microsoft Ads team and the project was about developing proof of concept models for "Ad-selection for CTS improvement".
  • Performed extensive analysis on various logs to get the right data for the modeling.
  • Developed a contextual bandit model using dropout as a Bayesian Approximation which will effectively reduce the latency in the pipeline with minimal impact on revenue.
  • Presented proof of concept by comparing its performance with the baseline Multi-Arm Bandit Model. There was an improvement of 7% in AUC.
Contextual BanditsBayesian StatisticsData AnalysisMachine LearningData Science

Education

Indian Institute of Technology, Guwahati

Bachelor of Technology

Jul 2018May 2022

Stackforce found 100+ more professionals with Machine Learning & Data Science

Explore similar profiles based on matching skills and experience

Rahul D. - AI Researcher | Stackforce