Ami Yip
Stream Initiative Manager @UBS
Signup · Get unlimited contacts
WORK HISTORY
Stream Initiative Manager @UBS
London, GB
Operating at the intersection of delivery, operating model, and workflow design across Asset Management Technology Digital Core Services (~300 FTE, 3 crews) — transforming fragmented execution into a system that is measurable, prioritised, and decision-driven. • Re-architected the Tech Mandatory portfolio from fragmented, anecdotal reporting into a single decision system, enabling leadership to prioritise based on risk, impact, and capacity — eliminating reliance on narrative-driven status updates • Shifted Agile from ritual to performance system, introducing throughput, cycle time, WIP, and flow health — making bottlenecks and delivery risk visible, and refocusing teams on outcomes over activity • Exposed systemic demand–capacity imbalance across crews, enabling more disciplined planning, reducing overcommitment, and improving predictability at stream level • Replaced slide-driven reporting with structured information flows, significantly reducing manual synthesis and turnaround time; applied Microsoft Copilot to compress reporting, communication, and analysis cycles • Began establishing an AI-enabled “memory layer” (early stage) — structuring operational knowledge (e.g. release notes, delivery signals) into reusable, machine-readable formats to support scalable insight generation • Built adoption mechanisms that changed behaviour, not just communication (Pizza & Present, Viva Engage, Innovation Hub), increasing visibility of work, surfacing ideas, and creating momentum for continuous improvement across DCS
EDUCATION
The University of Hong Kong
Postgraduate diploma of Information Technology(Part-time), Information Technology
Hong Kong Shue Yan University
Bachelor of Business & Administration, Finance concentration
ABOUT AMI YIP
Most organisations don’t struggle with AI because of technology.They struggle because no one trusts the output enough to act on it.AI can generate answers —but decisions are still made through meetings, opinions, and fragmented data.That gap — between intelligence and trust — is where I live.1. What I build- • AI-ready memory layers — structured, owned, human-validated knowledge • Decision intelligence — confidence-based voting, explicit trade-offs, captured reasoning • The change layer — what changed, why, and what decision was made • Real Agile — fast learning loops, not ceremonies • Reasoning-driven prioritisation — clear, defensible trade-offs2. How I think-AI doesn’t fail.It scales the system it operates in.If trust is low and knowledge is fragmented, AI amplifies noise.If systems are structured, governed, and human-validated, AI becomes a force multiplier.3. What I deliver-I build trusted systems where AI solves real problems — not just generates more output. • Human-in-the-loop ensures accuracy and accountability • Decisions are structured, validated, and reused • Systems continuously improve over timeFrom:more data, more noiseTo:clear decisions, measurable outcomes, continuous improvement4. How I deliver-Think big. Start small. Learn fast.I solve the trust problem step by step — starting with one real problem, validating accuracy through human-in-the-loop governance, and building continuous learning into the system before scaling further.That is how AI becomes trusted enough to drive real decisions — not just generate more output.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.