SUMMARY Trust Architects build trust in the AI and trust among the people who use it.
They make sure every system Avanade Next ships is safe, reliable and ready for production — and that the client's people are confident, capable and willing to adopt recent ways of working alongside it.
The role operates across two dimensions.
Trust in the AI: evaluation frameworks, accuracy measures, security controls, governance controls and human-in-the-loop safeguards.
Trust in the people: adoption design, capability uplift, ways of working and confidence in human-agent collaboration.
KEY RESPONSIBILITIES Building trust in the AI Design and run evaluations: define accuracy targets, build the harness, drive the system to target and evidence it.
Set governance, risk and security guardrails before build begins, and keep them current as the system evolves.
Embed guardrails, controls and human-in-the-loop review, including the feedback loop that improves the system after release.
Establish the evidence required for a system to move from prototype to production, and hold the release gate.
Own the regulatory and privacy position on the engagement — APRA, MAS, data residency and client-specific obligations.
Work shoulder to shoulder with FDEs so trust, evaluation and monitoring capability is built in from the outset.




Building trust in the people Design the adoption and change approach that builds confidence through transparency, education and demonstrated outcomes.
Help the client redesign ways of working and human-agent interaction so the system fits how people actually work.
Lead enablement so the client's people can and do use what has been built.
Drive uptake and user confidence after go-live, tracking active usage rather than reported satisfaction.
Support executive and Board confidence with clear governance, accountability and decision-making frameworks.
Contribute evaluation harnesses, guardrail patterns, governance frameworks and adoption assets to the APAC library.
KEY MEASURES Governance, guardrails and security controls in place before build starts on every engagement.
Target accuracy, safety and governance thresholds met and evidenced before production release.
Adoption above 70% active usage at 90 days post-go-live, with effective human-in-the-loop operation.
Full compliance with Avanade responsible AI standards and client regulatory obligations.
Reusable trust, evaluation and adoption assets contributed from every engagement.
undefined undefined undefined undefined undefined

📌 Trust Architect (New South Wales)
🏢 Avanade
📍 New South Wales