AIforce vs Agentforce and the new Salesforce developer role: notes from Dreamforce 2026

AIforce does not replace Agentforce. Developers are moving from writing every line of code to giving agents guidelines, reviewing their output and setting guardrails. And the real story of Dreamforce 2026 was MCP and skills, not a single vendor.

I'm Juan Manuel Garrido, founder of EGA Futura and co-founder of Vantegrate. On September 17 we streamed our weekly masterclass live from floor 61 of Salesforce Tower with Alba Rivas from Salesforce, right after the developer keynote. These are her three clarifications.

1. AIforce is not a rebrand of Agentforce

Start with the Headless Toolkit (formerly Headless 360, introduced at TrailblazerDX):

AIforce is the layer that uses those technologies to open Salesforce to any MCP client or agent, plus partnerships with agent vendors. Agentforce remains Salesforce's own AI. The idea that one replaces the other is, in Alba's words, a misconception that may have been poorly communicated.

2. From code to guidelines

Creating the object, writing queries, the class, the tests and deploying: according to Alba, that classic workflow doesn't have much time left. The role shifts and expands:

• Design: ask the agent for a solution that follows your guidelines, technologies and criteria.
• Review: check that what was built matches what was asked, because AI makes mistakes too.
• Guardrails: give the agent knowledge and limits.

Anyone can ask an agent to build an app. Not everyone can build one that works in production in an enterprise environment.

3. The Dreamforce of MCP and skills

The Claude relationship continues (there's already a Salesforce plugin for Claude), but OpenAI and NVIDIA were also on stage, and AIforce is meant to work with everyone. MCP is becoming a universal protocol, and business users use it too. Skills are written in Markdown and can be shared across a team.

Full recap of the session on Vantegrate Pulse (Spanish). Version with FAQ on EGA Futura Discover (Spanish).