3 min read
AI Engineer vs ML Engineer

AI Engineering

My thoughts

Having been in the data space for more than a decade, it’s been fascinating to see a new paradigm arrive and then split up.

There were Machine Learners.

Then there were Machine Learning people and Machine Learning Operations (ML Ops) people.

The Machine Learning people do the “academic” research to develop new models.

The MLOps people ensure that machine learning (ML) models are coded, tested, and deployed reliably and consistently.

Large Language Model (LLM) people then came into the picture.

From here you got a split into the people doing research to develop new models and the people using them.

So you saw the rise of the “Prompt Engineer”.

Fast forward a few years, and people started incorporating LLMs into daily business activities.

From this, you are seeing a rise of “AI Engineers.”

Per the article

An AI Engineer leverages pre-trained models and existing AI tools to enhance user experiences. Their primary focus is on the practical application of AI, rather than constructing models from the ground up. This approach distinguishes them from AI Researchers and ML Engineers, who are more concerned with developing new models or advancing AI theory.

In time, “AI Engineers” will also be responsible for AI Agents, which means connecting several LLMs to each other or to AI Agents.

As it stands, the Evolution of AI/ML Roles is:

  • ML Researchers: Focus on model development and theoretical advances
  • MLOps Engineers: Handle deployment and operational aspects
  • Prompt Engineers: Specialize in LLM interaction design
  • AI Engineers: Build practical applications using existing models

Workwise, AI Engineers typically work with:

  • API integrations
  • Vector databases
  • Retrieval-Augmented Generation (RAG)
  • AI Agent orchestration

While ML Engineers focus on:

  • Model architecture
  • Training pipelines
  • Model Optimization
  • Data preprocessing

We will likely see further role specialization as AI systems become more complex and interconnected.

Specifically for AI Agents, we might see:

  • AIOps Engineers who handle deployment and operational aspects of AI Agents
  • AI Agent Engineers who handle building the connections between AI Agents and the organizations they serve

This breakdown will represent a significant shift as it will require upskilling already existing AI Engineers and MLOps Engineers and providing new education to new developers.