AI Annotation Specialist

Posha builds cooking robots that live inside your kitchen and cook food for you completely autonomously. Posha robots use Artificial Intelligence enabled Computer Vision to consistently and reliably cook the best food for our users

See it in action:

Posha is a fast-growing Series-A funded startup backed by Tier-1 VCs including Accel Partners, Waterbridge Ventures.

What can I expect from this role?

Posha is one of the rare startups building cutting-edge deep tech consumer robotic products from India for the world. Our product sits at the intersection of AI, Software, Food and Design.

High-quality training data is the foundation of our AI systems.

As an AI Annotation Specialist, you'll be the critical bridge between our ML models and real-world performance—ensuring every piece of labeled data meets the exacting standards needed for autonomous cooking.

What will my work responsibilities look like if I join Posha?

In this role - you would:

• Monitor ML Model Health
  : Perform system health checks on deployed machine learning models, identifying data quality issues, annotation drift, or labeling inconsistencies that impact model performance.
• Quality Assurance & Validation
  : Validate annotations delivered by external labeling companies, ensuring complete alignment with our objectives for ingredient detection, cooking state tracking, and other vision tasks.
• Create Annotation Guidelines
  : Write clear, comprehensive annotation guidelines for new tasks (e.g., segmentation masks for vegetables, bounding boxes for ingredient identification, classification labels for cooking stages) that can be shared with labelling partners.
• Hands-on Collaboration
  : Work directly with the labelling company on initial batches of images to establish shared understanding, calibrate quality standards, and iron out edge cases before scaling.
• Iterative Refinement
  : Refine annotation rules and guidelines based on ongoing feedback from the ML team and labelling company - creating a continuous improvement loop for data quality.
• Cross-functional Pipeline Management
  : Actively collaborate with the kitchen-ops team to understand upcoming annotation needs, prioritise tasks, and build an efficient pipeline to get relevant cooking data annotated on time.

What makes you a match for us?

• Bachelor's degree
• 0-2 years of experience in data annotation, ML operations, or Computer Vision projects
• Strong attention to detail with the ability to spot inconsistencies in labeled data
• Basic understanding of computer vision concepts (object detection, segmentation, classification)
• Familiarity with annotation tools (e.g., CVAT, Label Studio, V7, Supervisely) is a plus
• Excellent written and verbal communication skills for creating guidelines and coordinating with external teams
• Comfortable working in an ambiguous, fast-paced startup environment where you'll need to define processes from scratch
• Passionate about quality and willing to dive deep into edge cases
• Bonus: Understanding of how ML models are trained and deployed

Working at Posha

• Twitter love from users
• How does Posha work
• Our Website
• Read Our Story

Why Posha?

• Building completely autonomous cooking robots requires thinking from first principles and building from ground up.
• Your work directly impacts whether our robots can distinguish between diced and julienned vegetables, or know when dal has reached the perfect consistency - these are novel, unsolved problems in computer vision.
• We believe that Posha is a generational company with the opportunity to alter the trajectory of how food is cooked inside homes today. This opportunity is parallel to how autonomous cars have made an impact.
• As the guardian of our training data quality, you'll be essential to making this vision a reality.