Workflow Engineering: The Next Essential Skill for AI Builders
Workflow engineering is emerging as a foundational discipline for AI engineers and users, sitting alongside context engineering as a "must-have" skill. While context engineering ensures agents have the right information at every step, workflow engineering focuses on designing repeatable, multi-step processes—often involving multiple LLM calls and tool integrations—that automate tasks traditionally performed by humans, such as customer support, legal research, and report generation. Unlike simple ReAct agents, which can be too unconstrained for complex tasks, workflow engineering enables the creation of structured, reliable, and context-aware agentic workflows. These workflows combine the adaptability of AI agents with the predictability of traditional automation, allowing for dynamic planning, tool use, and iterative improvement. Mastering workflow engineering empowers both technical and non-technical users to describe and automate sophisticated processes, ensuring AI delivers meaningful, production-grade outcomes
1. Workflow Engineering for AI Agents
We're all in on context engineering.
A related topic that imo is table stakes for every AI engineer/user: workflow engineering 🛠️
A lot of agent use cases revolve around automating work that otherwise a human would have to perform - customer support, legal research, report generation, unit testing, etc.
It obviously needs to be context-aware, but it also needs to be a repeatable multi-step process; there's usually a sequence of steps that you can describe that require multiple LLM calls to achieve a given outcome.
In most cases you can't fully trust a simple ReAct agent with tools - it is too unconstrained and doesn't necessarily fulfill the task at hand.
Both AI engineers and non-technical users need to get really good at describing these workflows, because that's how you get AI to meaningfully complete work for you instead of giving you back chat responses from the raw LLM API.
(If you go back to the model layer, you could even use RL/tuning to optimize/overfit to the workflow at hand, to ensure 100% accuracy vs. 80%)
I'm super bullish on workflow engineering; at the end of the day every AI builder is building specialized AI workflows in one way or the other. We're building document workflows at LlamaIndex, other companies are building customer support workflows, coding workflows, and more.
(Image from Philipp Schmid, there's another cool diagram by Dexter Horthy, also check out Tuana Çelik's post here: https://lnkd.in/gtZzf_3S)