Junior Project Manager (AI/ ML) - Asian market

📋 Description

• • Own the full project life-cycle for AI & ML initiatives that power Tsmg’s next-generation products across Asia-Pacific. From first customer workshop in Singapore to model deployment in Manila data-centres, you will translate ambitious business goals into executable road-maps, sprint plans and OKRs that deliver measurable ROI within 90-day cycles.
• • Build and nurture a virtual squad of data scientists, MLOps engineers, product owners and regional sales leads spread across Manila, Bangalore and Tokyo. Run daily stand-ups, weekly risk reviews and monthly steering committees that keep every timezone aligned while protecting maker-time for deep-work.
• • Translate fuzzy market requirements into crisp user stories, acceptance criteria and Definition-of-Done for machine-learning features such as real-time recommendation engines, multilingual NLP chatbots and computer-vision quality-control modules. Maintain a living backlog that balances model-accuracy improvements with latency, cost and regulatory constraints.
• • Design and enforce stage-gate governance that satisfies both agile velocity and enterprise compliance (ISO 27001, PDPA, GDPR). Maintain a RAID log that tracks model-drift, data-pipeline failures, GPU availability and ethical-AI risks; trigger pre-approved mitigation playbooks before issues hit customer SLAs.
• • Drive vendor selection and cloud-resource budgeting for AWS SageMaker, GCP Vertex and on-prem GPU clusters. Negotiate spot-instance pricing that cuts training cost by 30 % without sacrificing experiment reproducibility; present business-case deltas to Finance for quarterly budget re-allocation.
• • Champion responsible-AI practices: coordinate bias-testing, fairness audits and explainability reviews with external legal counsel; maintain model-cards and data-lineage documentation that pass regional regulator inspections in Thailand, Indonesia and the Philippines.
• • Create executive-facing dashboards in Jira & Tableau that translate technical metrics (F1-score, inference-latency, GPU-utilisation) into commercial KPIs (conversion uplift, call-deflection rate, OPEX savings). Tell the story behind the numbers so that non-technical VPs understand why retraining frequency matters to revenue.
• • Institutionalise continuous improvement: after each release conduct blameless retrospectives, update the AI PMO playbook, and mentor junior scrum-masters on hybrid agile/DevOps rituals. Your goal is to compress concept-to-cash cycle time by 25 % year-over-year while keeping team NPS above 80 %.
• • Stay ahead of the curve: attend APAC AI meet-ups, evaluate emerging auto-ML platforms and feed competitive insights back to strategy teams. Run quarterly “demo-days” where prototypes built during 10 % innovation time are pitched for seed funding and possible spin-off.