Insights
Practical thinking on building AI agents on Salesforce your team can govern and evolve.
Featured insight
Systems Architecture
How to Plan the Transition to Agentic Operations
A credible agentic architecture still needs a path from the systems you run today. Map coexistence, authority, costs and evidence before expanding deployment.
Latest insights
Systems Architecture
Salesforce Well-Architected: A Practical Guide to Better Architecture Decisions
Salesforce Well-Architected becomes useful when it changes a decision. It gives teams a way to test options, expose trade-offs, and retain an architecture rationale they can defend and revisit.
AI Governance
How AI Learns Changes What You Need to Prove
An introduction to the assurance challenges behind supervised, unsupervised and reinforcement learning.
Capability Ownership
Choosing a Salesforce Partner? Ask Its Past Clients What Lasted
A Salesforce partner award shows what Salesforce chose to reward, not what lasted at a client like you. Before you shortlist a partner, borrow the award criteria and check them where the work was done: with the partner's past clients.
Capability Ownership
If the Work Happens in Claude, What Is Salesforce For?
Salesforce wants a role in agent work even when the interface belongs to another vendor, and it says its capabilities can be used whole or in part alongside what a company already has. If you already use Salesforce, the question is what this adds to what you have. If you do not, the question is what would justify bringing it in.
Systems Architecture
How to Choose an Agent-Design Method for Governed AI Workflows
Four published methods can turn a governed-workflow brief into an agent architecture. The right one satisfies the project's non-negotiable constraints without hiding its gaps or dependencies.
AI Governance
What Did Go-Live Actually Prove?
A go-live acceptance proves that a solution was delivered and can operate. It does not prove that the organisation has gained a measurable operational advantage — or built the capability to sustain it. Once the accepted system starts deciding and acting, that unproven interval stops waiting quietly to be measured.
Capability Ownership
From Systems of Record to Governed Systems of Action
AI adoption is spreading faster than the operational change that would make it pay. This article argues that the governed workflow, not the platform implementation, should become the unit around which the next phase of transformation is commissioned, measured and controlled.
AI Governance
AI Governance: Choosing the Right Controls for Your Business
How learning methods shape the risks, evidence and oversight your AI systems need.