Data Quality Analyst | Human Archive
Location: Bangalore | Salary: Rs 22,000 30,000/month
Employment Type: Full-time | Experience: Freshers welcome

We are Human Archive

Human Archive is an AI company capturing real human experience through first-person (egocentric) video.

What do we do?

We are building the infrastructure and data layer for Physical AI. Physical AI means robots and AI systems that can see, move and act in the real world, not just answer questions on a screen.

To learn, these systems need huge amounts of high-quality, real-world data. In the early 2020s, companies like Scale AI built the data layer that helped language models (like ChatGPT) grow. We are doing the same thing for Physical AI. We collect the data, process it, check its quality and deliver it to the companies building these systems.

This is an early and fast-growing field, and you will be part of it from the ground up.

What does the role look like?

You will be the final layer of checking and processing before the data reaches our customers. Every video and label passes through your hands before it goes out. If something is wrong, you are the one who catches it.

This means your work directly decides the quality of what we deliver, and that quality decides how well AI models learn. It is a role with real responsibility, not just repetitive work.

What you will do here

- Watch and review video: Go through first-person video clips carefully and understand what is happening in each one.
- Trim clips: Cut videos to exact start and end timestamps so each clip shows one clear action or task.
- Label data: Mark the actions, objects and context in each clip, following clear guidelines.
- Quality check (QC): Check your own work and the work of others for mistakes before it moves forward.
- Flag problems: Spot and report footage that is unclear, blurry,



incomplete or does not follow the rules.
- Process data for delivery: Make sure the final data is clean, correct and ready to send to customers.
- Meet daily targets: Keep a steady output while keeping accuracy high.

What you will learn

- How to judge data: Over time you will build a strong sense of what good data looks like and what bad data looks like, and why it matters.
- How to make decisions using rules: You will learn to apply guidelines and principles to make clear, consistent decisions, even when a case is not obvious.
- How AI training data is made: You will see the full journey, from raw video recorded in the field to the final dataset a customer uses.
- Attention to detail: You will train your eye to notice small things that most people miss.

These skills are in high demand as AI grows, and they will be useful wherever you go next.

Who we are looking for

- Someone who is patient and pays close attention to detail
- Someone who can follow guidelines carefully but also thinks for themselves when something looks wrong
- Someone who is honest about mistakes and wants to improve
- Someone who is curious about AI and how it learns

No technical background needed. We give full training before you start.

Who can apply

- Class 12 pass minimum, graduation preferred
- Age 18 and above
- Freshers, students and college dropouts welcome
- Positive English reading and writing (you will read guidelines and write labels in English)

Schedule

Monday to Saturday | 8 hours/day | Flexible start time

Work Culture

- We don't track when you come in. We care about your work, not the clock. Come in at a time that works for you and complete your 8 hours.
- Casual dress code. Wear what you are comfortable in.
- No hierarchy. Everyone is approachable, from team leads to founders. If you have a question or an idea, just ask.

📌 AI Data Annotator (Bengaluru)
🏢 Human Archive
📍 Bengaluru