Harsh Sharma

Data Scientist

Gurugram, Haryana, India3 yrs 10 mos experience

Key Highlights

  • Expert in AI and machine learning applications.
  • Proven track record in healthcare data analytics.
  • Strong background in developing scalable AI solutions.
Stackforce AI infers this person is a Data Scientist specializing in AI-driven solutions for Healthcare and Marketing sectors.

Contact

Skills

Core Skills

Machine LearningData ScienceComputer Vision

Other Skills

Data AnalysisPredictive ModelingData ProcessingStatistical AnalysisNLPAutomated SystemsDeep LearningMentoringFace RecognitionAngularData AnalyticsTableauAmazon Web Services (AWS)Software DevelopmentDjango

About

I am a Data Scientist with a strong background in AI, deep learning, and data-driven decision-making. My expertise spans across machine learning, GenAI, NLP, and large-scale data analysis, with a focus on building intelligent systems that drive innovation. What I Do: ๐Ÿ”น Develop AI-powered solutions to solve real-world challenges in healthcare, finance, and automation ๐Ÿ”น Build and optimize deep learning models for predictive analytics and computer vision applications ๐Ÿ”น Design and deploy scalable data pipelines and AI architectures ๐Ÿ”น Apply advanced statistical and data science techniques to derive actionable insights Iโ€™m passionate about using AI for impact-driven solutions and always eager to discuss cutting-edge innovations in machine learning, data science, and AI ethics. Letโ€™s connect and explore the future of AI together!

Experience

3 yrs 10 mos
Total Experience
1 yr 1 mo
Average Tenure
5 mos
Current Experience

Epam systems

Data Scientist

Dec 2025 โ€“ Present ยท 5 mos ยท Gurugram, Haryana, India ยท Hybrid

Zs

Data Scientist

Feb 2025 โ€“ Dec 2025 ยท 10 mos ยท Gurugram, Haryana, India ยท Hybrid

  • Led data-driven analyses to identify and refine target Healthcare Providers (HCPs) and patient segments,
  • increasing campaign reach efficiency by 30% and improving patient engagement outcomes.
  • Extracted, cleaned, and integrated complex healthcare datasets from multiple sources, reducing data
  • processing time by 25% and ensuring high-quality inputs for predictive modeling and insights.
  • Built and deployed tree-based models (XGBoost) to predict HCP prescribing behavior and patient
  • treatment pathways with 92% accuracy; leveraged SHAP for interpretability, surfacing the top 5 drivers
  • that directly informed targeting strategy.
  • Partnered with cross-functional teams to translate model-driven insights into actionable
  • recommendations, improving client decision-making and driving a 15% uplift in ROI on commercial
  • strategy initiatives.
Data AnalysisPredictive ModelingData ProcessingMachine LearningStatistical AnalysisData Science

Consultadd inc

SDE - Data Science

Jul 2023 โ€“ Jan 2025 ยท 1 yr 6 mos ยท Pune District, Maharashtra, India

  • Developed an ๐—ก๐—Ÿ๐—ฃ-๐—ฏ๐—ฎ๐˜€๐—ฒ๐—ฑ ๐—ถ๐—ป๐—ณ๐—น๐˜‚๐—ฒ๐—ป๐—ฐ๐—ฒ๐—ฟ ๐˜๐—ฎ๐—ด๐—ด๐—ฒ๐—ฟ, which increased targeted advertising ๐—ฒ๐—ณ๐—ณ๐—ถ๐—ฐ๐—ถ๐—ฒ๐—ป๐—ฐ๐˜† ๐—ฏ๐˜† ๐Ÿด๐Ÿฌ%, generating $80,000 in revenue.
  • Led the development of an ๐—ฎ๐˜‚๐˜๐—ผ๐—บ๐—ฎ๐˜๐—ฒ๐—ฑ ๐—ฝ๐—ฎ๐—ฟ๐˜๐˜€ ๐—ฑ๐—ฒ๐˜๐—ฒ๐—ฐ๐˜๐—ถ๐—ผ๐—ป ๐˜€๐˜†๐˜€๐˜๐—ฒ๐—บ for generator maintenance, improving reporting accuracy by 40%.
  • Designed and implemented ๐—™๐—– (๐—™๐˜‚๐—น๐—น๐˜† ๐—–๐—ผ๐—ป๐—ป๐—ฒ๐—ฐ๐˜๐—ฒ๐—ฑ) ๐—ฐ๐—น๐—ถ๐—ฝ ๐—บ๐—ผ๐—ฑ๐—ฒ๐—น๐˜€ to enhance action recognition in retail store environments, improving the ability to identify and ๐—ฝ๐—ฟ๐—ฒ๐—ฑ๐—ถ๐—ฐ๐˜ ๐—ฐ๐˜‚๐˜€๐˜๐—ผ๐—บ๐—ฒ๐—ฟ ๐—ฏ๐—ฒ๐—ต๐—ฎ๐˜ƒ๐—ถ๐—ผ๐—ฟ๐˜€.
  • Utilized ๐—ž๐—ก๐—ก (๐—ž-๐—ก๐—ฒ๐—ฎ๐—ฟ๐—ฒ๐˜€๐˜ ๐—ก๐—ฒ๐—ถ๐—ด๐—ต๐—ฏ๐—ผ๐—ฟ๐˜€) algorithms in ๐—ฎ๐—ฐ๐˜๐—ถ๐—ผ๐—ป ๐—ฟ๐—ฒ๐—ฐ๐—ผ๐—ด๐—ป๐—ถ๐˜๐—ถ๐—ผ๐—ป systems for retail store environments, optimizing the tracking and recognition of customer actions for better insights and operations.
  • Collaborated with cross-functional teams to integrate machine learning models into production, ensuring ๐—ต๐—ถ๐—ด๐—ต ๐—ฎ๐—ฐ๐—ฐ๐˜‚๐—ฟ๐—ฎ๐—ฐ๐˜† and reliability of the implemented solutions.
  • Spearheaded ๐—”๐—œ-๐—ฑ๐—ฟ๐—ถ๐˜ƒ๐—ฒ๐—ป ๐—ถ๐—ป๐—ป๐—ผ๐˜ƒ๐—ฎ๐˜๐—ถ๐—ผ๐—ป to streamline processes, enhance decision-making, and optimize client-facing applications.
NLPAutomated SystemsMachine LearningComputer Vision

Dsdl kiet

Data Science and ML Mentor

Sep 2021 โ€“ Oct 2022 ยท 1 yr 1 mo

  • ๐—ง๐—ฎ๐˜‚๐—ด๐—ต๐˜ ๐—ฐ๐—ผ๐—น๐—น๐—ฒ๐—ด๐—ฒ ๐˜€๐˜๐˜‚๐—ฑ๐—ฒ๐—ป๐˜๐˜€ the fundamentals of ๐——๐—ฎ๐˜๐—ฎ ๐—ฆ๐—ฐ๐—ถ๐—ฒ๐—ป๐—ฐ๐—ฒ ๐—ฎ๐—ป๐—ฑ ๐——๐—ฒ๐—ฒ๐—ฝ ๐—Ÿ๐—ฒ๐—ฎ๐—ฟ๐—ป๐—ถ๐—ป๐—ด, covering key topics like ๐—ฑ๐—ฎ๐˜๐—ฎ ๐—ฝ๐—ฟ๐—ฒ๐—ฝ๐—ฟ๐—ผ๐—ฐ๐—ฒ๐˜€๐˜€๐—ถ๐—ป๐—ด, ๐—บ๐—ฎ๐—ฐ๐—ต๐—ถ๐—ป๐—ฒ ๐—น๐—ฒ๐—ฎ๐—ฟ๐—ป๐—ถ๐—ป๐—ด ๐—ฎ๐—น๐—ด๐—ผ๐—ฟ๐—ถ๐˜๐—ต๐—บ๐˜€, ๐—ป๐—ฒ๐˜‚๐—ฟ๐—ฎ๐—น ๐—ป๐—ฒ๐˜๐˜„๐—ผ๐—ฟ๐—ธ๐˜€, ๐—ฎ๐—ป๐—ฑ ๐—ฑ๐—ฒ๐—ฒ๐—ฝ ๐—น๐—ฒ๐—ฎ๐—ฟ๐—ป๐—ถ๐—ป๐—ด ๐—ณ๐—ฟ๐—ฎ๐—บ๐—ฒ๐˜„๐—ผ๐—ฟ๐—ธ๐˜€.
  • ๐——๐—ฒ๐˜€๐—ถ๐—ด๐—ป๐—ฒ๐—ฑ ๐—ฎ๐—ป๐—ฑ ๐—ฑ๐—ฒ๐—น๐—ถ๐˜ƒ๐—ฒ๐—ฟ๐—ฒ๐—ฑ ๐—ฒ๐—ป๐—ด๐—ฎ๐—ด๐—ถ๐—ป๐—ด ๐—น๐—ฒ๐—ฐ๐˜๐˜‚๐—ฟ๐—ฒ๐˜€ with practical examples to bridge the gap between theoretical knowledge and real-world applications in data science.
  • ๐—š๐˜‚๐—ถ๐—ฑ๐—ฒ๐—ฑ ๐˜€๐˜๐˜‚๐—ฑ๐—ฒ๐—ป๐˜๐˜€ ๐˜๐—ต๐—ฟ๐—ผ๐˜‚๐—ด๐—ต ๐—ต๐—ฎ๐—ป๐—ฑ๐˜€-๐—ผ๐—ป ๐—ฝ๐—ฟ๐—ผ๐—ท๐—ฒ๐—ฐ๐˜๐˜€ and assignments, ensuring they gained proficiency in tools like ๐—ฃ๐˜†๐˜๐—ต๐—ผ๐—ป, ๐—ง๐—ฒ๐—ป๐˜€๐—ผ๐—ฟ๐—™๐—น๐—ผ๐˜„, ๐—ž๐—ฒ๐—ฟ๐—ฎ๐˜€, ๐—ฎ๐—ป๐—ฑ ๐—ฃ๐˜†๐—ง๐—ผ๐—ฟ๐—ฐ๐—ต.
  • ๐— ๐—ฒ๐—ป๐˜๐—ผ๐—ฟ๐—ฒ๐—ฑ ๐˜€๐˜๐˜‚๐—ฑ๐—ฒ๐—ป๐˜๐˜€ on developing and deploying machine learning models, fostering critical thinking and problem-solving skills.
  • Received positive feedback for ๐—ฐ๐—ฟ๐—ฒ๐—ฎ๐˜๐—ถ๐—ป๐—ด ๐—ฎ๐—ป ๐—ถ๐—ป๐˜๐—ฒ๐—ฟ๐—ฎ๐—ฐ๐˜๐—ถ๐˜ƒ๐—ฒ ๐—น๐—ฒ๐—ฎ๐—ฟ๐—ป๐—ถ๐—ป๐—ด ๐—ฒ๐—ป๐˜ƒ๐—ถ๐—ฟ๐—ผ๐—ป๐—บ๐—ฒ๐—ป๐˜ that encouraged collaboration and ๐—ฒ๐—ป๐—ต๐—ฎ๐—ป๐—ฐ๐—ฒ๐—ฑ ๐˜€๐˜๐˜‚๐—ฑ๐—ฒ๐—ป๐˜ ๐—ฒ๐—ป๐—ด๐—ฎ๐—ด๐—ฒ๐—บ๐—ฒ๐—ป๐˜.
Data ScienceDeep LearningMentoring

Lens corporation

Computer Vision Research Intern

Sep 2021 โ€“ Apr 2022 ยท 7 mos ยท Gurugram, Haryana, India ยท Hybrid

  • Researched and implemented multiple face recognition algorithms, including ๐——๐—ฒ๐—ฒ๐—ฝ๐—™๐—ฎ๐—ฐ๐—ฒ, ๐—”๐—ฟ๐—ฐ๐—™๐—ฎ๐—ฐ๐—ฒ, ๐—™๐—ถ๐˜€๐—ต๐—ฒ๐—ฟ๐—™๐—ฎ๐—ฐ๐—ฒ, ๐—ฎ๐—ป๐—ฑ ๐—˜๐—ถ๐—ด๐—ฒ๐—ป๐—™๐—ฎ๐—ฐ๐—ฒ.
  • Conducted comparative analysis of ๐˜๐—ฟ๐—ฎ๐—ฑ๐—ถ๐˜๐—ถ๐—ผ๐—ป๐—ฎ๐—น ๐˜ƒ๐˜€. ๐—ฑ๐—ฒ๐—ฒ๐—ฝ ๐—น๐—ฒ๐—ฎ๐—ฟ๐—ป๐—ถ๐—ป๐—ด-based models to evaluate accuracy and efficiency.
  • Developed and optimized ๐—ด๐—ฒ๐—ป๐—ฑ๐—ฒ๐—ฟ ๐—ฐ๐—น๐—ฎ๐˜€๐˜€๐—ถ๐—ณ๐—ถ๐—ฐ๐—ฎ๐˜๐—ถ๐—ผ๐—ป models, improving recognition accuracy across diverse datasets.
  • Analyzed performance metrics to enhance ๐—บ๐—ผ๐—ฑ๐—ฒ๐—น ๐—ฟ๐—ผ๐—ฏ๐˜‚๐˜€๐˜๐—ป๐—ฒ๐˜€๐˜€, ๐—ณ๐—ฒ๐—ฎ๐˜๐˜‚๐—ฟ๐—ฒ ๐—ฒ๐˜…๐˜๐—ฟ๐—ฎ๐—ฐ๐˜๐—ถ๐—ผ๐—ป, ๐—ฎ๐—ป๐—ฑ ๐—ฟ๐—ฒ๐—ฎ๐—น-๐˜„๐—ผ๐—ฟ๐—น๐—ฑ ๐—ฎ๐—ฝ๐—ฝ๐—น๐—ถ๐—ฐ๐—ฎ๐—ฏ๐—ถ๐—น๐—ถ๐˜๐˜†.
  • Explored ๐—ฑ๐—ฒ๐—ฒ๐—ฝ ๐—น๐—ฒ๐—ฎ๐—ฟ๐—ป๐—ถ๐—ป๐—ด techniques to improve ๐—ณ๐—ฎ๐—ฐ๐—ถ๐—ฎ๐—น ๐—ฟ๐—ฒ๐—ฐ๐—ผ๐—ด๐—ป๐—ถ๐˜๐—ถ๐—ผ๐—ป under varying lighting and occlusion conditions.
Face RecognitionDeep LearningComputer Vision

Education

KIET Group of Institutions

Bachelor of Technology - BTech โ€” Computer Science

Jan 2019 โ€“ Jan 2023

Delhi Public School Ghaziabad

12th โ€” PCM

Jul 2017 โ€“ Jul 2019

Mount Carmel School, Muradnagar

10th

May 2017 โ€“ Present

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

Explore similar profiles based on matching skills and experience