As foundation models continue to improve, I think AI engineering is starting to look far more like distributed systems engineering.
The difficult part usually is not the model itself - it is everything around it:
- Orchestration
- Retries
- Queues
- Workflow state
- Observability
- Evaluation
- Scaling
A production AI workflow can very quickly become:
- Retrieval
- Multiple LLM/tool calls
- Async processing
- Validation
- Downstream systems
At that point, you are dealing with classic system problems rather than just prompting.