Senior ML Engineer at Yubo

Yubo is the Social Discovery app designed to help users make new friends and connect online by eliminating likes and follows, fostering genuine interactions.

As Yubo scales, Machine Learning has become a core production layer, powering critical systems in safety, recommendations, and product optimization. This role is unique due to the scale and diversity of our data, including massive volumes of images, text, and real-time user interactions across millions of users worldwide. Key challenges include:

• Content moderation (image, text, behavior)

• Recommendation systems and user engagement optimization

• Behavioral detection and trust & safety models

• Emerging use cases like dynamic pricing and growth optimization

The current ML stack is evolving, with legacy models needing integration, inconsistent lifecycle management, and a develop, deploy, and forget approach. As ML usage expands, there is huge untapped potential requiring a more robust and scalable ML operating model.

Your Responsibilities

ML Systems & Delivery

• Deliver end-to-end ML use cases (recommendation systems, safety algorithms, moderation models, etc.)

• Design, train, evaluate, and deploy ML models where appropriate

• Balance speed of delivery with robustness and reliability

ML Lifecycle & Reliability

• Drive improvements across the ML lifecycle (training, deployment, monitoring, iteration)

• Establish KPIs and monitoring standards to track model performance over time

• Ensure continuous alignment with product and safety objectives

• Improve the reliability and observability of production ML systems

Platform & Standardization

• Contribute to the evolution of our ML platform through tools, workflows, and reusable components

• Help establish and promote scalable ML engineering practices across teams

• Contribute to improving self-service ML capabilities where relevant

Legacy & Advanced Use Cases

• Take ownership of legacy models and realign them with current business needs

• Improve, retrain, and integrate them into modern pipelines

• Contribute to best practices around LLM usage in moderation, recommendation, and other product domains

• Explore and implement advanced ML approaches where relevant

Cross-functional Collaboration

• Partner with Data Engineering, MLOps, Backend Platform, and Product teams

• Act as a bridge between ML, platform, and business stakeholders

• Provide technical leadership on ML initiatives and help improve engineering practices across teams

• Share knowledge and help raise the overall ML maturity of the organization

Technical Environment / ML Scope

• ML frameworks: PyTorch

• Languages: Python (data stack)

• Core topics: Neural networks, LLMs, data sampling

• Use cases: Recommendation systems, safety algorithms, moderation

• Ecosystem: Data Engineering, MLOps pipelines, Backend Platform

Who You Are

• You have 5+ years of experience in ML/Data, including work on large-scale datasets (hundreds of gigabytes)

• Strong expertise in modern ML frameworks (TensorFlow or PyTorch or JAX)

• Highly proficient in Python (data ecosystem)

• Strong knowledge of neural networks and practical experience with LLM-based systems

• Understand ML systems end-to-end (data → training → deployment → monitoring)

• Strong experience in production ML systems, not just research

• Strong product sense and ability to align models with business needs

• Pragmatic and impact-driven, balancing experimentation with delivery

• Strong pedagogy to explain complex ML topics clearly

• Operate well under ambiguity and structure complex problems

• Drive technical decisions and influence stakeholders through expertise and collaboration

In Your First 6 Months

• Deliver an end-to-end ML use case to validate impact and execution

• Audit and improve lifecycle management of existing ML models

• Refactor and retrain at least one legacy safety model

• Build reusable ML components that improve development efficiency across teams

• Improve monitoring and observability of production ML systems

• Contribute to the evolution of our ML platform and engineering practices

• Help move critical ML systems toward more reliable and measurable long-term operation