Senior Software Engineer 1, ML

People Inc. is focused on building a next-generation product discovery platform that connects shoppers with their desired products. As a Senior Software Engineer for personalization, you will design and develop the recommendation algorithm that enhances user experience by personalizing their product feeds based on individual preferences.

Responsibilities:

• Design and build the core personalization engine using user-saved product data as behavioral signals
• Develop multi-signal recommendation models that incorporate brand affinity, product category, color palette, fit/sizing signals, price sensitivity, and trends
• Implement and evaluate a range of approaches including collaborative filtering, content-based filtering, and hybrid neural architectures
• Build and maintain product embedding models that capture rich semantic similarity across the retailer feed catalog
• Develop cold-start strategies to generate high-quality recommendations for new users with limited save history
• Design and maintain robust pipelines to ingest, normalize, and enrich product feeds from thousands of retail partners
• Collaborate on a unified product taxonomy and attribute extraction layer that standardizes inconsistent retailer data into coherent features (category, color, material, fit, etc.)
• Leverage NLP and computer vision techniques to extract attributes from unstructured product descriptions and images
• Partner with the data engineering team to maintain data quality, freshness, and catalog coverage at scale
• Build and own the ranking and re-ranking layer that assembles each user's personalized feed in real time
• Develop and tune multi-objective ranking that balances relevance, novelty, diversity, and business goals (e.g., promoted/sponsored retailer partnerships)
• Implement feedback loops that continuously update user preference models based on implicit signals (saves, clicks, dwell time, shares)
• Build A/B testing solutions to rigorously evaluate ranking and recommendation changes against key engagement metrics
• Own production systems. Debug issues across indexing, retrieval, ranking, and serving layers
• Create clear documentation for pipelines, models, APIs, and system design
• Contribute to best practices for ML systems, API design, and scalable infrastructure
• Stay current with advancements in recommendation, ranking, and personalization systems and apply them where they make practical impact

Requirements:

• Bachelor's degree in Computer Science, Engineering, or a related field
• 5+ years of ML engineering experience focused on recommendation systems, personalization, or search ranking with hands-on depth in collaborative filtering, matrix factorization, content-based, and hybrid neural approaches
• Proven experience designing, training, and deploying embedding models and vector retrieval (e.g., Milvus, Pinecone) for product or content similarity at catalog scale
• Production experience serving real-time, low-latency ML predictions and managing the full model lifecycle — training, deployment, versioning, and monitoring — on cloud ML platforms such as AWS SageMaker or GCP Vertex AI (including Vertex AI Pipelines)
• Rigorous experimentation discipline: experiment design, A/B and multivariate testing, and the analytical ability to translate model results into clear product and business decisions
• Extensive backend engineering with strong proficiency in Python and modern ML frameworks (PyTorch, TensorFlow, or JAX), plus working knowledge of Node.js and TypeScript
• Experience designing large-scale data and feature pipelines using Apache Kafka, Spark, Beam, Airflow, or Flink for streaming ingestion, transformation, and feature engineering
• Applied NLP and/or computer vision experience extracting structured attributes (category, color, material, fit) from unstructured product descriptions and imagery
• Strong API and infrastructure foundations: REST and GraphQL design with secure auth (OAuth/JWT), Git-based workflows, containerization with Docker and Kubernetes, and production observability with Grafana, Kibana, and APM tooling
• Curiosity and pragmatism around emerging AI, particularly LLMs and modern retrieval/ranking techniques, with a track record of bringing new approaches into real production use cases
• Strong written and verbal communication, able to explain technical tradeoffs to both technical and non-technical stakeholders, with a data-driven approach to problem solving
• Backend and API development using Python, FastAPI, Node.js, and TypeScript
• Search and indexing using Elasticsearch for relevance, retrieval, and query optimization
• Event driven architecture and streaming using Apache Kafka
• Vector search and embeddings infrastructure using vector databases such as Milvus or Pinecone
• Cloud and infrastructure using Google Cloud Platform or Amazon Web Services with containerization via Docker and orchestration through Kubernetes