ML career Q&A: ML Engineer vs. AI Engineer
Celebrating 500 newsletter subscribers!
I got a question from a recent talk on Analytics Vidhya, and thought to share the response I gave + elaborate some more. The newsletter also allows me to share screenshots and examples. The question from an attendee was “What is the difference between ML Engineer roles and AI Engineer roles?”
Here’s some context: Recently, Generative AI has been very hot, and new roles have been created to work on Generative AI, RAG (retrieval augmented generation, and so on, using tools such as LangChain, various vector databases, and working with various LLMs such as those provided by OpenAI, or those on HuggingFace.
There are speculations that there will be an entirely new role called AI Engineers that will appear and will focus specifically on tasks such as those.
If you’re looking for such a job related to GenAI, then I will repeat the advice I gave in Machine Learning Interviews: look at the job description. Here is a real life example:
At my company, people working with Generative AI have the following titles:
Software engineer
Data Scientist
Machine Learning Engineer
At other companies, I know a few of these roles that have switched to working on Generative AI projects for some of their time, while working on non-Generative-AI projects like before in the other half of their time.
Just like the title, “Data Scientist”, many, if not most companies aren’t going to immediately replace their job titles, even if they are doing different tasks, such as data engineering or MLOps.
In addition, I find that companies are hesitate to switch titles completely, in case they need to hire generalists who can work on Generative AI but also other tasks. For example, if they hire for “Software Engineer — Generative AI”, then a successful candidate can also improve other parts of the software related to GenAI, but not only the narrow portion of GenAI.
The point being, if you want to work with LLMs, Generative AI, and related technologies, MLE, and even DS are still quite viable. You have to look at the job descriptions; here are some examples that I found.
Note: I did not include the URL links to these jobs, since some of them have already been taken down by the time I’m sending this newsletter.
Examples: Machine Learning Engineer (MLE)
Quora:
Capgemini:
Examples: Data Scientist
IBM:
Examples: Software Engineer
Elastic:
Mozilla:
In summary, same with the job titles Data Scientist, Applied Scientist, Machine Learning Engineer, MLOps engineer… what you will be doing depends on the team, company, and projects. The best way to guess what will be your responsibility is to look carefully at the job descriptions, and to ask your recruiter or hiring manager. If you’re interested in Generative AI, don’t be hesitant to seek those projects in job titles where other job competitors aren’t looking. If you’re interested in MLOps, recommender systems, reinforcement learning, etc… don’t discount a certain job title before you take a look.
Are you interested in more Machine Learning career advice and tips but don't want to wait until the next newsletter? You can check out my book, Machine Learning Interviews for a comprehensive starter resource.
Thank you to those who have reached out to ask how they can support this newsletter so I can continue to write more career advice for ML job seekers. While this newsletter is free, there is a paid pledge option on Substack for those who would like to support it.
And if you found this issue useful, I would really appreciate it if you could share it with your friends and colleagues, or restack it here on Substack.