Data Analyst I

Elsevier; Mumbai; On-site; INR 260,000 - 380,000; Full time

Job summary

Elsevier is seeking detail-oriented AI Annotators to support identity verification datasets, including document fraud, deepfake, PAD, and face recognition tasks. You will label and validate diverse data, improve model accuracy, and contribute to robust data pipelines.

The role requires attention to detail, ability to perform repetitive tasks with high accuracy, and comfort using annotation tools. Training will be provided, with opportunities to work closely with ML engineers and data scientists.

Qualifications

• Strong attention to detail and accuracy in labeling.
• Ability to perform repetitive tasks reliably.
• Basic understanding of computer vision and ML concepts.
• Experience with web-based annotation tools (training provided).
• Clear written and verbal communication to describe edge cases.

Responsibilities

• Annotate data across documents, images, video, and audio following project guidelines.
• Label document elements (names, photos, signatures) and flag tampering.
• Annotate manipulated regions in deepfake and image tamper datasets.
• Tag liveness indicators and spoof artifacts in PAD datasets.
• Draw bounding boxes, facial landmarks, and expression labels for face recognition tasks.
• Apply time-based labels in video for frame-level PAD/deepfake detection.
• Document edge cases and suggest improvements to guidelines.
• Review peer annotations for consistency and quality.
• Collaborate with ML engineers and data scientists to improve dataset quality and tools.

Skills

Attention to detail

Repetitive task focus

CV/ML concepts basics

Annotation tools experience

Clear communication

Tools

Web-based annotation tools

Job description

About the Team

You’ll join a collaborative engineering team building scalable, high-quality software solutions across backend, frontend, and cloud. The team works closely with cross‑functional partners, values knowledge sharing and continuous learning, and encourages everyone to contribute ideas and improvements while balancing technical excellence with delivery.

About the Role

We are seeking detail‑oriented AI Annotators to support the development of secure and fair identity verification systems. Your work will involve labeling and validating datasets used in document fraud detection, deepfake detection, presentation attack detection (PAD), and face recognition. The annotations you provide will directly improve model accuracy, fairness, and resilience.

Responsibilities

• Data annotation across multiple formats (documents, images, video, audio) following project guidelines.
• Identify and label document elements (e.g., names, photos, signatures, holograms) and flag potential tampering.
• Annotate manipulated regions in deepfake and image tamper datasets.
• Tag liveness indicators and spoof artifacts in PAD datasets.
• Draw bounding boxes, facial landmarks, and expression labels for face recognition tasks.
• Apply time‑based labels in video for frame‑level PAD/deepfake detection.
• Document edge cases and suggest improvements to annotation guidelines.
• Review peer annotations, check consistency, and ensure quality standards.
• Collaborate with ML engineers and data scientists to improve dataset quality and tools.

Requirements

• Attention to detail and strong observational skills.
• Ability to focus on repetitive tasks while maintaining accuracy.
• Basic understanding of computer vision/ML concepts (training provided).
• Comfortable using web‑based annotation tools (training provided).
• Clear written and verbal communication skills to describe edge cases.