Responsible AI
Human accountability, bounded actions, evaluation, and rollback.
CANADIAN PUBLIC-SECTOR TECHNOLOGY PARTNER
Amidship is a Canadian, principal-led software company for public-sector teams that need important technology designed, built, operated and evolved without separating senior judgment from the work itself.
The principal shaping the architecture stays close to implementation, production decisions and change over time.
WHY THE MODEL MATTERS
Amidship is a Canadian SME headquartered in Pickering, Ontario. The delivery model is intentionally principal-led: discovery, product decisions, architecture, implementation direction and final technical review do not disappear into separate sales, solutioning and delivery layers after a contract is signed.
That does not mean every task is performed by one person. It means the senior person accountable for the system remains accountable as the work moves from decision to implementation to operation.
BUILD. OPERATE. EVOLVE.
Some engagements should end in a clean handoff. Others should become long-running technology relationships. The operating model should match the system, not a consulting template.
Map the workflow, operating constraints, existing systems, risks and the decisions that actually matter.
Choose the architecture, controls and delivery shape before implementation momentum makes those decisions expensive to revisit.
Implement the software, integrations, automation and operating controls with inspectable progress and clear decision ownership.
Monitor, support, recover, upgrade and keep the production system understandable after launch.
Change the system as policies, workflows, models, integrations and organizational needs change.
MODERNIZE WITHOUT STARTING OVER
A modern agentic system should not require an organization to discard every deterministic automation, API, business rule or legacy workflow that already works. Amidship designs around those investments first.
Where the boundary is sound, existing RPA processes, APIs, line-of-business systems and human approvals can remain execution assets inside a newer orchestration layer. Replace what creates risk. Keep what still earns its place.
PRODUCTION IS PART OF THE WORK
Amidship’s client work includes a long-running production software relationship spanning application development, integrations, infrastructure, maintenance, staged releases and ongoing production support. The point is not that every public-sector system should be operated the same way. It is that our delivery history extends beyond prototypes and first releases.
Read the production software storyDetailed named past-performance material is available for qualified procurement and diligence contexts where client permission allows.
ENTERPRISE EXPERIENCE, DIRECTLY INVOLVED
Founder and principal Rida Al Barazi brings more than two decades of product and software architecture experience across founder-led companies, Wave and Salesforce. His current work includes product and software architecture inside a large enterprise environment, with a current technical focus on agentic systems, tool use, tool-using agent workflows, integration, state, permissions, evaluation and human control.
Salesforce and Wave are Rida’s key-person experience, not Amidship corporate past performance. That distinction stays explicit in our procurement material.
About RidaTrust and delivery
Responsible AI, security, accessibility, privacy, and a concise capability statement are available for teams that need to understand how Amidship works before a mandate begins.
Human accountability, bounded actions, evaluation, and rollback.
Least privilege, separated environments, secrets discipline, and third-party controls.
Semantic structure, keyboard access, visible focus, contrast, and testing.
A concise supplier profile covering capabilities, delivery model, relevant experience, and contact details.
PROCUREMENT AND DILIGENCE
The website explains how Amidship works. The capability statement and controlled past-performance material carry the attributable procurement detail: corporate facts, key-person experience, selected evidence, delivery model and approved client references.