Applied Scientist III - Recommendation Systems

Overview

Build and deploy large-scale recommendation systems and agentic AI systems that power Glance's personalized lock screen and live entertainment experiences.

What you'll do

• Design and develop large-scale recommendation systems using advanced ML, statistical modeling, ranking algorithms, and deep learning.
• Build and operate machine learning models on diverse, high-volume data sources for personalization, prediction, and content understanding.
• Develop rapid experimentation workflows to validate hypotheses and measure real-world business impact.
• Own data preparation, model training, evaluation, and deployment pipelines in collaboration with engineering counterparts.
• Monitor ML model performance using statistical techniques; identify drifts, failure modes, and improvement opportunities.
• Build and experiment with agentic AI systems that autonomously observe model performance, trigger experiments, tune hyperparameters, improve ranking policies, or orchestrate ML workflows with minimal human intervention.
• Apply LLMs, embeddings, retrieval-augmented architectures, and multimodal generative models for semantic understanding, content classification, and user preference modeling.
• Design intelligent agents that can automate repetitive decision-making tasks, e.g., candidate generation tuning, feature selection, or context-aware content curation.
• Explore reinforcement learning, contextual bandits, and self-improving systems to power next-generation personalization.
• Collaborate with Designers, UX Researchers, Product Managers, and Software Engineers to integrate ML and GenAI-driven features into Glance's consumer experiences.
• Contribute to Glance's ML/AI thought leadership, including blogs, case studies, internal tech talks, and industry conferences.
• Interface with stakeholders across Product, Business, Data, and Infrastructure to align ML initiatives with strategic priorities.

What you'll need

• Bachelor's or Master's in Computer Science, Statistics, Mathematics, Electrical Engineering, Operations Research, Economics, Analytics, or related fields.
• 6+ years of industry experience in ML/Data Science, ideally in large-scale recommendation systems or personalization.
• Experience with large-scale ML and recommendation systems (collaborative filtering, ranking models, content-based approaches, embeddings).
• Experience with classical ML and deep learning techniques across NLP, sequence modeling, RL, clustering, and time series.
• Experience deploying ML workflows/models in production systems.
• Experience with big data processing (Spark, distributed data systems) and cloud computing.
• Experience designing end-to-end ML solutions from prototype to production.
• Expertise with algorithms in NLP, Reinforcement Learning, Time Series, and Deep Learning, applied on real-world datasets.
• Proficient in Python and comfortable with statistical tools (R, NumPy, SciPy, PyTorch/TensorFlow, etc.).
• Strong experience with the big data ecosystem (Spark, Hadoop) and cloud platforms (Azure, AWS, GCP/Vertex AI).
• Comfortable working in cross-functional teams.
• Excellent communication skills with the ability to simplify complex technical concepts.

Nice to have

• PhD.
• Experience with LLMs, retrieval systems, generative models, or agentic/autonomous ML systems.
• Building or experimenting with LLMs, generative models, and agentic AI workflows (e.g., autonomous evaluators, self-improving pipelines, automated experiment agents).
• Familiarity with privacy-preserving ML and identity-less ecosystems (especially on iOS and Android).

Details

• Location: Bangalore.