Français
Deutsch
The democratisation nobody predicted
In 2023, building an AI agent required Python skills, knowledge of LLM APIs, and familiarity with orchestration frameworks. In 2025, a motivated Business Analyst can deploy an operational agent in a few hours, without writing a single line of code. This transformation fundamentally changes organisational dynamics.
The no-code agentic ecosystem in 2025
| Platform | Positioning | Ideal for |
|---|---|---|
| Gumloop | AI-first, visual workflow | BA/PM, complex AI workflows |
| Flowise | Open source, RAG specialist | Enterprise chatbots, knowledge base |
| Relevance AI | Business agents | Sales, marketing, support |
| Stack AI | Enterprise no-code LLM | Large organisations |
| Langflow | Visual LangChain | Technical teams wanting a visual interface |
What Business Analysts can now do independently
The boundary between "business professional" and "developer" is blurring. Here's what a BA can build today without code:
- Research agents: browse the web, synthesise information and write reports
- Document processing agents: extract, classify and route incoming documents
- Communication agents: reply to emails, create tickets, send notifications
- Analysis agents: query databases in natural language and generate visualisations
New skills of the augmented BA
From "how to code" to "how to orchestrate"
The key skill is no longer a programming language's syntax, but the ability to design orchestration workflows: identifying steps, defining inputs/outputs, handling exception cases.
From "what to ask devs" to "what to tell agents"
Prompt engineering becomes a core BA skill. Precisely formulating an instruction for an AI agent, with the right level of detail and appropriate constraints — that's the new "writing a user story".
"The 2026 Business Analyst will be judged on their ability to orchestrate AI agents as much as their ability to write functional specifications."