Everything we have published on agentic engineering.
AI does not make Salesforce developers obsolete. It moves their work up a level: define outcomes, write specs, review with expertise and orchestrate teams of agents. After years of unclear messaging and tool chaos, Salesforce showed what the future might look like.
The FDE job starts before anyone builds anything: deciding which agents are worth building, and deciding how you will measure success or failure once they run.
Our steps and learnings while building a semi-autonomous AI pipeline that takes a GitHub ticket and converts it into an implemented pull request.
Personalisation in software usually means removing friction. But friction, discoverability and serendipity make us better engineers. How I built AI skills that put learning back in the loop.
See how AI agents like Claude Code speed up manual QA on Salesforce projects, from pulling JIRA tickets to writing bug reports and retest evidence.
Two pricing announcements on the same effective date will reshape AI development costs. Architecture decisions made now define the AI expenses for the next two years.
Building an AI proof-of-concept is the easy part. Most initiatives stall when it's time to run those solutions on real data, under security, compliance, and operational constraints. The FDE model exists to close that gap.
Fast to create. Harder to secure, scale, and maintain. Why "build your own CRM with AI" looks like a $300B savings on paper and a costly software business in practice.