Hassan M.

AI projects don't slip for technical reasons. They slip because nobody defined what "good enough" means for a model that's right 85% of the time, the eval process was an afterthought, and scope kept growing every time someone said "can it also...". That's what I fix. I'm a PMP-certified delivery manager with 11 years running technology projects — most recently a cybersecurity transformation programme at Emirates Group, and before that enterprise IT delivery at Orange Business with full margin accountability. Upwork Inc. hired me directly to support staffing for a large-scale AI/ML programme, assessing around 15,000 specialist profiles across three markets. 100% Job Success Score, 5.0 across 21 reviews. Send me an invite and I'll reply with a short personalised Loom — how I'd structure your build, the risks I'd hit first, and what week one looks like. What I do on AI projects: • Define acceptance criteria and eval gates for probabilistic systems — so "it works" means something specific • Run delivery across ML engineers, data, product and vendors, keeping decisions moving • Manage the dependencies that actually kill AI timelines: data access, API limits, model changes, annotation throughput • Rescue builds that have drifted from a demo into a stalled six-month project • Turn a vague "we want to use AI for X" into scope, owners and a plan someone can execute Tools: ClickUp, Jira, Asana, Monday, Notion, Microsoft Project, Trello, Slack, Teams and Google Workspace. Tell me what you’re building, what is currently stuck and what successful delivery should look like. I’ll help you create a practical path forward.