Amidship

Operational AI · Product architecture · Integration

Operational AI that actually ships.

Amidship helps organizations turn complex workflows into governed AI-enabled products and services—from discovery and architecture through implementation and evaluation.

Discuss a workflow See our work

Independently verifiable credentials

2015

Founded

Amidship Inc.

Registered

Government of Canada supplier

Direct verification link pending

Proof link pending

20+ years

Principal experience

Rida Al Barazi

What we solve

Where ambitious AI work meets operating reality.

  1. 01

    Operational workflows with too much manual coordination

    Find where AI or automation belongs, then move one bounded workflow toward operation.

  2. 02

    Agentic systems that need real controls

    Design tools, permissions, state, approvals, observability, and evaluation as part of the system—not as an afterthought.

  3. 03

    AI-enabled products and services

    Turn an ambiguous opportunity into tested software with a credible production path.

  4. 04

    Complex modernization and integration

    Connect AI to existing applications, data, teams, and operating constraints.

Why Amidship

Architecture judgment, product discipline, and principal continuity.

Senior accountability stays on the work.

Rida leads discovery, architecture, consequential technical decisions, and final review.

Workflow before model.

Product and system fit come before provider or model selection.

Enterprise judgment. Product speed.

Salesforce and Wave are Rida’s key-person experience—not Amidship corporate past performance.

Small by design, disciplined in delivery.

Specialist resources can be added to a defined scope. Amidship does not present a salaried delivery bench.

Selected proof

Inspect the work, not the adjectives.

Evidence classes and maturity are stated on every item. Corporate work, key-person experience, and R&D are not interchangeable.

R&D-02 Product R&D · current maturity

Agentify

An active Rails R&D platform exploring stateful agents, tools, channels, retrieval, evaluation, and human review.

Active R&D; no customer or production-adoption claim

Inspect the repository (opens in a new tab)

Evidence gate: The former Amidship service-business platform is intentionally omitted until product history, metrics, and publication permission are verified.

Delivery model

Accountable from first question to final review.

Amidship is principal-led. Scope-specific specialists may be added when the work calls for them; their access, responsibility, and review path are explicit.

Trust by design

Controls belong in the delivery system.

These are public practices, baselines, and commitments—not certifications or audit claims.

TRUST-01Practice

Responsible AI

Human accountability, bounded agent actions, evaluation, transparency, and rollback.

Full public baseline follows in Phase 1
TRUST-02Baseline

Security

Least privilege, environment separation, secrets discipline, and explicit third-party controls.

Full public baseline follows in Phase 1
TRUST-03Commitment

Accessibility

Semantic structure, keyboard access, visible focus, contrast, responsive text, and testing.

Full public baseline follows in Phase 1
Rida Al Barazi, founder and accountable principal at Amidship
Rida Al Barazi · Accountable principal

Principal continuity

Senior judgment stays close to the work.

Rida Al Barazi brings more than 20 years of product and software architecture experience to discovery, consequential technical decisions, and final review.

His experience at Salesforce and Wave is key-person experience. It demonstrates the perspective he brings; it is not presented as Amidship corporate past performance.

Independent profile and writing GitHub

For government evaluators

AI and software delivery for accountable public-sector environments.

Amidship Inc. is a Canadian corporation based in Pickering, Ontario, Canada, founded in 2015 and registered as a Government of Canada supplier.

Best-fit mandates include bounded operational AI, agentic workflows, prototypes, architecture and integration, and evaluation or governance. No public-sector delivery reference or certification is implied.

Procurement inquiry

Start with the operating problem

Have an operational problem that AI might actually improve?

Discuss a workflow

Share the workflow, operating context, and what is getting in the way. No mandatory budget questionnaire.