Project Manager, Applied AI

Role overview

Lead complex, multilingual data programs that power the next generation of AI language systems. This role sits at the intersection of operations and applied AI, coordinating global contributor networks and steering large-scale annotation, evaluation, and safety initiatives from kickoff through retrospective.

Responsibilities

- Drive end-to-end delivery of AI data projects, including scoping, guideline development, execution, quality review, and post-mortem analysis. - Manage pipelines for multilingual data collection across audio, text, and image modalities, alongside large language model evaluation efforts such as RLHF, supervised fine-tuning, ranking studies, and safety testing. - Track and report KPIs spanning throughput, accuracy, inter-annotator agreement, gold-set performance, cost-per-task, and worker productivity. - Run quality assurance loops, perform root-cause analysis when quality dips, deliver corrective training to contributor pools, and maintain real-time dashboards that surface bottlenecks. - Coordinate with distributed teams of linguists and data specialists to meet service-level expectations around localization nuance and linguistic accuracy. - Bridge technical requirements into clear, actionable instructions for non-technical annotators, and channel field insights back into guideline updates and model fine-tuning strategies.

Requirements

- Three to five or more years of project management experience within AI or machine learning data operations. - Working knowledge of large language model training lifecycles (pre-training, supervised fine-tuning, RLHF) and evaluation methodologies such as human-in-the-loop review and red teaming. - Advanced proficiency in spreadsheets plus the ability to write SQL queries to extract and analyze operational data. - Demonstrated success applying Agile, Scrum, or Kanban to manage complex, multi-stream workflows. - Strong written communication skills with the ability to produce unambiguous guidelines for multilingual audiences.

Nice to have

- Fluency in a second language. - Hands-on experience with annotation platforms (for example, Scale AI or SuperAnnotate) and project management tools such as Jira. - Academic or professional background in machine learning engineering, computer science, data science, or formal project management training.