Hybrid profile
Where Salesforce and infrastructure are one head, not two
I'm not a Salesforce developer who knows servers, nor a sysadmin who knows Salesforce. I solve cross-cutting problems that a single specialist — on either side — would handle more slowly, less well, or not at all: both layers' decisions are made at the same time, by the same person, before the usual back-and-forth between teams even starts. The evidence cited below comes from training and personal projects; each page states its type and period.
Three cases where the combination changes the outcome
- 01
Heroku ↔ Salesforce data migration, with a field mobile app
The problem — An insurer acquires a competitor: the acquired CRM's data has to flow into Salesforce (one-way), and field agents need a mobile app that reads and writes against the same org.
A single specialist
A Salesforce profile stops at the Apex REST layer and hands off the Heroku infrastructure — integration decisions are made without knowing the constraints on the other side. An infrastructure profile deploys the Heroku app, but designs neither the External ID Apex trigger nor the Matching Rule that stop duplicates at the source.
What I deliver
Both: the Apex REST endpoints, the Node.js/PostgreSQL app that consumes them, Heroku Connect configured object by object, two isolated environments linked by a pipeline, right down to the deployment runbook. The staging environment existed before the first line of Apex was written — not after.
Where it matters — Insurance and finance: reconciling two customer books where a duplicate isn't an inconvenience but an orphan file and a compliance risk.
- 02
DevOps discipline applied to Salesforce releases
The problem — Shipping Salesforce changes repeatedly without breaking production, with auditable traceability on every go-live.
A single specialist
A Salesforce profile can code the feature, but deploys via manual Change Set, with no documented rollback procedure. A DevOps profile can build a pipeline, but doesn't know the pitfalls of deploying Salesforce metadata — no native undo, dependency order between components.
What I deliver
Real DevOps discipline — automated validation of every Pull Request against the org, server-to-server JWT Bearer authentication, a private key wiped from the runner after every run, a documented SFDX rollback procedure — applied to how Salesforce releases are delivered, not just how they're coded.
Where it matters — Any regulated sector — healthcare, energy, finance: every production change must be traceable and reversible, not just functional.
- 03
An integration's access control, designed on the Salesforce side
The problem — A Salesforce integration exposes REST endpoints: which user calls them, with which permissions, and which data can it reach?
A single specialist
A Salesforce developer ships the endpoint relying on an existing, overly permissive profile. An infrastructure team configures server-side authentication — but doesn't know which Permission Set, which CRUD/FLS check or which Sharing Rule actually limits access on the Salesforce side.
What I deliver
I design the integration's permissions on the Salesforce side — a dedicated integration user, an explicit Permission Set, CRUD/FLS checks, WITH SECURITY_ENFORCED — together with server-side authentication, in the same sprint. Access is scoped before go-live, not three tickets later.
Where it matters — Healthcare and luxury: integrations that move patient data or product provenance — where poorly scoped access isn't a technical detail but a breach of trust.
What the combination removes
- The back-and-forth between the Salesforce team and the infra/DevOps team on every integration decision.
- Decisions made on one side without knowing the constraints of the other — a trigger designed without knowing how the infra calls it, a pipeline designed without knowing Salesforce's deployment limits.
- The gap between "the API is audited" and "the control is in place": both happen in the same sprint.
Hiring for a cross-cutting role?
If the role touches both Salesforce and infrastructure — integration, migration, traceable go-live, deployment automation — that's exactly where this profile saves a team time.