Harsh J.

Operations Associate

Delhi, India0 mo experience

Key Highlights

Developed state-of-the-art deep learning model for yield prediction.
Improved operational efficiency through data analysis and workflow design.
Led community outreach and social media initiatives.

Stackforce AI infers this person is a Data Scientist with expertise in Deep Learning and Business Analysis.

Contact

Skills

Core Skills

OptimizationBusiness AnalysisDeep LearningMachine LearningComputer VisionCommunity OutreachSocial Media Communications

Other Skills

Data VisualizationInteger ProgrammingJavaMicrosoft ExcelNumPyOpenCVPandas (Software)People ManagementProblem SolvingProject ManagementPublic RelationsPublic SpeakingPyTorchPython (Programming Language)R (Programming Language)

Experience

0 mo

Total Experience

Average Tenure

Current Experience

Zs

Associate

Jul 2025 – Present · 11 mos · Pune, Maharashtra, India · On-site

OptimizationPython (Programming Language)R (Programming Language)

Newgen software

Associate Business Analyst

Mar 2025 – Jul 2025 · 4 mos · Noida, Uttar Pradesh, India · On-site

Collaborated with cross-functional teams to analyze business requirements and support software solution development for enterprise clients.
Improved operational efficiency by gathering and interpreting data to identify process inefficiencies and designing workflows using Newgen’s low-code platform.

Business AnalysisMicrosoft ExcelProblem Solving

Tech mahindra

Research Intern

Jul 2024 – Aug 2024 · 1 mo · United States · Remote

Deepfake Detection, Multimodals
collaborating on research and Working on deepfake detection system for Multimodlities (Audio-Visuals).
Worked with Pre-Trained models for feature extarction task
on Deepfake Datasets.

PyTorchPython (Programming Language)Pandas (Software)NumPyOpenCVDeep Learning+3

University of virginia

Research Intern

May 2024 – Sep 2024 · 4 mos · United States · Remote

Developed Terraformer, a novel deep learning model integrating CNN, SegNet, and Mask R-CNN for precise crop mapping and yield prediction, achieving a state-of-the-art RMSE of 0.34 in yield prediction for the USA, surpassing existing models’ performance of 0.37 RMSE, with practical applications published in academic papers.

PyTorchPython (Programming Language)Deep LearningComputer VisionMachine LearningTime Series Analysis