Data Scientist (Computer Vision & Geospatial AI)

About Us

We are looking for a skilled and detail-oriented Data Scientist to play a crucial role in our computer vision projects. This individual will be responsible for the full data lifecycle, from collection and cleaning to annotation and analysis, ensuring that our machine learning models have the highest quality data to learn from. The ideal candidate has a strong background in data science with a specific focus on preparing datasets for computer vision tasks.

This role will work alongside ML Engineers and Data Engineers to design datasets, define evaluation metrics, source high-quality data, and extract actionable insights from AI models.

Key Responsibilities

• Data Sourcing & Collection: Identify, source, and manage large-scale datasets, including satellite imagery and multi-angle product photos.
• Data Annotation & Labelling: Oversee and perform the meticulous annotation and labelling of images with bounding boxes and class labels for object detection tasks. This includes ensuring consistency and accuracy across the dataset.
• Data Cleaning & Preprocessing: Clean, transform, and preprocess raw data to make it suitable for machine learning models. This involves handling missing data, normalizing features, and correcting errors.
• Data Augmentation: Design and implement data augmentation strategies to increase the size and diversity of the training dataset, improving model robustness and generalization.
• Exploratory Data Analysis (EDA): Perform detailed EDA to understand the characteristics of the data, identify trends, identify potential biases, and provide insights that guide model development.
• Develop statistical and ML models for size estimation, orientation, and movement patterns.
• Quality Assurance: Develop and implement quality control protocols for the annotated data to ensure a high level of accuracy and reliability.
• Collaboration: Work closely with Machine Learning Engineers and Domain experts to define data requirements, provide data-driven insights, and troubleshoot data-related issues during the model training and evaluation process.
• Work with ML Engineers to select and evaluate models (YOLO, DETR, Mask R-CNN, etc.).
• Define KPIs and create evaluation frameworks (precision/recall, IoU, orientation accuracy).
• Analyse model errors, identify failure cases, and propose improvements.
• Generate dashboards/reports to communicate insights to stakeholders.

Required Skills & Experience

• 3+ years of experience as a Data Scientist, ideally in computer vision or geospatial AI.
• Strong skills in Python (Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn).
• Experience with image datasets and libraries like OpenCV, Pillow.
• Familiarity with deep learning frameworks (PyTorch/TensorFlow) to run experiments.
• Hands-on experience with Satellite & geospatial data (GDAL, Rasterio, Shapefiles) is a plus.
• Strong grasp of statistical modelling, hypothesis testing, and evaluation design.
• Experience with large-scale dataset preparation, annotation, and sourcing workflows.
• Proficiency with SQL and data pipelines.
• Excellent communication skills — able to translate technical results into business insights.

Nice to Have (Preferred)

• Familiarity with multi-object tracking and trajectory analysis.
• Knowledge of segmentation and fine-grained classification problems.
• Exposure to data visualization tools (Tableau, Power BI, Dash, Plotly).
• Experience working with annotation teams and quality assurance processes.
• Experience working with AIS data and open-source maritime datasets

What We Offer

• Opportunity to work on impactful AI projects
• Collaboration with a strong ML/AI engineering team.
• Flexible working setup.
• Competitive compensation and growth opportunities.

Job Type: Full-time

Pay: ₹600,000.00 - ₹700,000.00 per year

Benefits:

• Health insurance
• Provident Fund

Application Question(s):

• Are you open to work from office for Nashik location?

Experience:

• Data Scientist (Computer Vision & Geospatial AI): 3 years (Required)

Work Location: In person