Milena T.

Overview I build AI automations and agents that take repetitive work off a team’s plate, and I check the output before anyone relies on it. Example of my work My Invoice Automation Pipeline in n8n reads incoming invoices in Email, uses the Claude API to extract the key fields like invoice number, amount, and due date into structured data, sorts results into Google Sheets by confidence score, and sends a scheduled email digest. Nobody has to type invoice details by hand. I use Claude as a coding assistant and debug as I go, so I can explain how each part works. More projects are on my GitHub (Milena-Tanui) and in my portfolio. AI Training and Evaluation Scale AI Team Lead: I led 30+ contributors, including multilingual teams, through 1,000+ annotation and evaluation tasks, covering onboarding, training, quality checks, and delivery. LLM evaluation: I rated model outputs on tool-calling tasks for training usefulness, realism, and correctness, and built a multi-dimensional rating framework. Annotation: 10+ projects across text, image, and video, using CVAT, Roboflow, Labelbox, and Label Studio. Operations I’m an Operations Lead at Mobility for Climate Policy Research Institute, where I coordinate teams with workplans and SOPs, manage follow-up, and oversee the organization's operations. Working with me I send clear updates, flag problems early, and work more than 30 hours a week.