> For the complete documentation index, see [llms.txt](https://jamiewen00.gitbook.io/ai-engineering-handbook/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://jamiewen00.gitbook.io/ai-engineering-handbook/roles-and-teams/ai-engineer.md).

# AI Engineer

Software engineers who master LLMs. They build AI applications and ship features.

Product-minded. Design-aware. Full-stack. Focused on business outcomes.

<figure><img src="/files/kwzRDZUmQkCI0A4WVqnb" alt="ai engineer"><figcaption></figcaption></figure>

## Responsibilities

### Application Development

* Build AI apps using foundation models
* Create LLM interfaces
* Integrate AI into products via APIs

### Prompt Engineering

* Write effective prompts
* Build context systems
* Manage memory for conversations

### Evaluation Systems

* Select models
* Benchmark progress
* Detect production issues
* Build evaluation frameworks

### Infrastructure Operations

* Optimise inference speed and cost
* Manage model serving
* Monitor performance
* Handle compute clusters

### Data Engineering

* Curate datasets
* Process unstructured data
* Implement retrieval systems
* Control data quality

## Essential Skills

### AI-Specific Knowledge

* Prompt engineering techniques
* Model selection strategies
* Model evaluation methold
* Model fine-tuning approaches

### Product Development

* UX development for AI interfaces
* Experimentation workflows
* Business outcome measurement
* Problem-solving with AI

## ML Engineer to AI Engineer

From Model Training to Orchestration

* **Before**: Train models from scratch
* **After**: Orchestrate existing foundation models

From Data-First to Product-First

* **Before**: Collect data, then build
* **After**: Build fast, validate, then optimise

From Research to Application

* **Before**: Design algorithms
* **After**: Build user experiences

## Product Engineer to AI Engineer

AI-Specific Knowledge

* **Learn**: LLM capabilities and limits
* **Develop**: Prompt engineering skills
* **Understand**: Model selection

Data Handling Skills

* **Expand**: Unstructured data processing
* **Master**: Context retrieval
* **Build**: Data quality workflows