How about the career of AI data annotation engineer?

This article mainly explains the specific situation and related career prospects of the ai data labeling engineer.
Author: Ren Congcong
Date: April 18, 2023

Data is the soul of AI. The corresponding data in the natural world are somewhat inaccurate, messy, and invalid, and need to be collected, sorted, classified, and processed by humans. Among them, in the AI ​​industry environment, data annotation is a data source for AI, and it is also a learning path that AI cannot or cannot do without.

1. Employment prospects

Demand

insert image description here

employment trends

insert image description here
It can be seen that 163-3-37+22+217=362, the market demand has increased by more than 3.5 times in recent years

2. Personal opinion

It is a transitional position in AI technology, and the development prospect is currently relatively optimistic and rising. It is suitable for fresh graduates to choose to enter the industry. However, most of this kind of work is manual labor, without high technical accumulation, and you need to choose other positions, with limited room for advancement.

3. The advantages and disadvantages of data labeling engineers (referred to as labeling)

advantage

1. It can be used as an entry into other positions of AI (it can pave the way for engineering positions in machine vision, opencv, image recognition, visual inspection, etc.) 2. Simple and no need to use your brain
.

shortcoming

1. Low technical content
2. Manual labor
3. Low salary

4. Types of data labeling

1. Picture annotation

Such as: a large number of defect picture labeling work required by visual inspection equipment (different types of defects and types)

2. Audio annotation

Such as: Classify and label an audio (Mandarin, Cantonese, Shandong dialect, Wu dialect, etc.)

3. Text annotation

Such as: Emotional labeling (emotions, anger, sorrow, joy) on a sentence

5. Conditions required for entry

1. Master the use of certain labeling software, such as picture labeling labelme (an open source picture labeling software)

2. Master certain industry knowledge, such as nlp, chat-gpt, opencv and other technical principles.

3. High school, technical secondary school and above can be employed.

Guess you like

Origin blog.csdn.net/hj960511/article/details/130221530