Maximize Your Impact with TP
Welcome to TP, a global hub of innovation and empowerment, where we redefine the future. With a remarkable 10 billion annual revenue and a global team of 500,000 employees serving 170 countries in over 300 languages, we lead in intelligent, digital-first solutions.As a globally certified Great Place to Work in 72 countries, our culture thrives on diversity, equity, and inclusion. We value your unique perspective and believe that your talent is the missing piece that completes our vision for a brighter, digitally driven tomorrow.
The Opportunity
TP.ai Data Services partners with the world's leading technology companies to design, build, and scale trustworthy AI systems. We deliver high-quality, multi-modal data solutions that fuel innovations in Generative AI, Computer Vision, NLP, RLHF, and Trust & Safety. Our mission is to ensure AI systems are not only high-performing but also responsible, reliable, and ready for global scale.We are seeking a Data Scientist (AI/ML Ops) to take a central role in shaping how AI models are operationalized and brought to life at scale. In this role, you will design, implement, and optimize advanced data pipelines, model evaluation frameworks, and monitoring processes that ensure AI systems are trained and deployed with accuracy, efficiency, and compliance. You'll apply your expertise in data science and ML operationalization to create automated feedback loops, establish robust evaluation metrics, and drive continuous improvement across the AI lifecycle.This is a high-impact opportunity to work on frontier AI initiatives with global reach, partnering directly with leading Big Tech enterprises. You will collaborate closely with Data Engineers, AI Researchers, IT, and Product Teams to translate complex business challenges into data-driven solutions, while influencing best practices in Responsible AI and MLOps. Beyond building pipelines and models, you'll help define the standards for how safe, scalable, and trustworthy AI systems are delivered worldwide.
The Responsibilities & Duties
- Design data preprocessing pipelines for multi-modal datasets (text, audio, video, images).
- Engineer features and transformations for model training, fine-tuning, and evaluation.
- Apply data augmentation, synthetic data generation, and bias mitigation techniques.
- Develop and evaluate ML models for classification, ranking, clustering, and anomaly detection tailored to operational use cases.
- Expertise in power analysis, significance testing, and advanced methods such as bootstrapping and Bayesian inference.
- Create offline/online evaluation metrics (accuracy, F1, precision/recall, latency, fairness metrics) to track model performance.
- Support RLHF pipelines by integrating human feedback into model optimization loops.
- Collaborate with engineering teams to integrate models into production pipelines.
- Design automated monitoring, retraining triggers, and continuous improvement loops to maintain model quality in dynamic environments.
- Build dashboards and reporting frameworks for real-time monitoring of model and data health.
- Define labeling strategies, taxonomy structures, and annotation guidelines for supervised learning.
- Partner with annotation platforms to ensure accuracy, throughput, and SLA compliance.
- Perform statistical quality checks (inter-annotator agreement, sampling-based audits).
- Partner with clients, product teams, and operations leads to translate business problems into data/ML solutions.
- Work with Legal/Privacy teams to ensure datasets meet regulatory and ethical AI standards.
- Contribute to internal best practices, reusable frameworks, and knowledge bases for AI/ML Operations.
The Qualifications
- Master's or Ph.D. in Computer Science, Data Science, Applied Mathematics, or related fields.
- 5+ years of experience in ML/AI data science with a focus on operational deployment.
- Track record of delivering ML models into production environments at scale.
- Programming: Python (NumPy, Pandas, Scikit-learn, PyTorch/TensorFlow), SQL.
- ML Techniques: Supervised/unsupervised learning, NLP, CV, RLHF.
- MLOps Tools: MLflow, Kubeflow, SageMaker, Vertex AI, or similar.
- Data Handling: Spark, Dataflow, or equivalent big data frameworks.
- Visualization & Reporting: Tableau, Looker, or similar BI tools.
- Experience with Generative AI model evaluation (LLMs, diffusion models).
- Knowledge of Responsible AI frameworks (bias/fairness metrics, explainability).
- Exposure to Trust & Safety operations or high-volume data environments.
Pre-Employment Screenings
By TP policy, employment in this position will be contingent on your successful completion and passage of a comprehensive background check, including global sanctions and watch list screening.
Important | Policy on Unsolicited Third-Party Candidate Submissions
TP does not accept candidate submissions from unsolicited third parties, including recruiters or headhunters. Applications will not be considered, and no contractual association will be established through such submissions.
Diversity, Equity & Inclusion
At TP, we are committed to fostering a diverse, equitable, and inclusive workplace. We welcome individuals from all backgrounds and lifestyles and do not discriminate based on gender identity or expression, sexual orientation, race, religion, age, national origin, citizenship, disability, pregnancy status, veteran status, or other differences.
