Search & Recommendation Algorithm Engineer

About the Role

We are looking for a Search & Recommendation Algorithm Engineer to lead the research, development, and architectural design of algorithms for our overseas AIOS platform. This position focuses on creating precise, multi-lingual, family-oriented search and recommendation systems for global household users. Expertise in large-scale online platforms, machine learning modeling, and integrated search-recommendation systems is essential.

Key Responsibilities

• Develop end-to-end search algorithms incorporating user intent understanding, multi-language query parsing, multi-path recall, fine ranking, and re-ranking, aiming to enhance click-through rates, conversion rates, and user satisfaction.
• Create personalized recommendation systems based on a two-tiered user profile model representing both household groups and individual members. Address issues such as conflicting multi-user interests and cold-start scenarios across different ranking stages (recall, coarse, fine, re-ranking).
• Drive unified system architecture by merging real-time recommendation and search functionalities. Standardize content representation, user behavior modeling, and ranking methodologies to support bidirectional information exchange between search intent and recommendation modules.
• Enhance global multi-lingual and vector-based search capabilities by optimizing cross-lingual semantic retrieval, multi-language query processing, and harmonizing voice and text search vectors for diverse international markets.
• Construct complete data feedback loops encompassing exposure, click, conversion, and feedback metrics, and conduct ongoing A/B testing for continuous algorithm improvement.

Qualifications & Requirements

• Possess a Bachelor's degree or higher (Master’s preferred) in Computer Science, Mathematics, Software Engineering, or related disciplines.
• Have a minimum of 5 years of R&D experience specializing in Search, Recommendation, or NLP algorithm development, with documented large-scale system deployments.
• Proficient in machine learning models including Two-Tower Recall, DeepFM, Wide & Deep models, DIN, and multi-objective ranking frameworks.
• Skilled in Python programming and experienced with machine learning frameworks such as PyTorch or TensorFlow.
• Hands-on experience with search technologies and vector databases like Elasticsearch, Faiss, and Milvus.
• Strong background in feature engineering, sequential user behavior modeling, and using both real-time (e.g., Flink) and offline (e.g., Spark) computation frameworks.
• Familiar with key online metrics such as CTR, CVR, Recall, NDCG, and proficient in designing and analyzing statistical A/B experiments.

Preferred Expertise

• Experience working on search or recommendation algorithms tailored for TV platforms, smart home devices, or streaming media services.
• Advanced knowledge in multi-lingual natural language processing, cross-lingual retrieval systems, retrieval augmented generation (RAG) with large language models, and household multi-user behavior modeling.