Software Engineer II - Recommendations

Klaviyo•Boston, MA

16h•Onsite

About The Position

The Recommendations Platform Team is responsible for developing and deploying machine learning-based recommendation systems at scale, and building out the foundation for new use cases for technologies such as embedding-based similarity search to power agentic workflows. We are evolving to act as a layer of product intelligence, making sense of customer data in order to personalize messages with the right item at the right time across multiple channels. Our systems span large-scale data pipelines and querying workflows, batch training and inference, and low-latency online retrieval and ranking systems. We are also building the experimentation, tracking, and measurement capabilities needed to evaluate recommendation quality and business impact over time.

Requirements

2+ years of professional software engineering experience with a focus on backend and distributed systems at scale; you have a proven track record working on production services and optimizing for latency, reliability, and operability as well as business requirements.
Proficient in Python and open to working in other languages
Comfortable with cloud-native architectures (AWS preferred) and container orchestration (e.g., Kubernetes); you manage infrastructure and CI/CD pipelines as a core part of your development process.
Experience in data-driven decision making and A/B testing —you can define (or are interested in learning how to) how to instrument experiments, read and interpret results, and ensure learnings are folded back into system design.
Comfortable designing and querying data models in relational, analytical, and NoSQL datastores (e.g., Postgres, MySQL, data warehouses, Redis, vector databases).
Feel at home with modern DevOps practices (CI/CD, monitoring, alerting) and how to apply them to architect large-scale data and recommendation systems.
Track record of owning features end-to-end —from initial technical design and implementation through rollout, monitoring, and sustained iteration.
Excellent technical collaborator and communicator: you can clearly articulate complex technical trade-offs to both technical peers and non-technical partners, and you work effectively to drive alignment across ML Engineers, Software Engineers, PMs, and other teams.
You are a self-starter who has actively experimented with AI in work or personal projects and are excited to responsibly explore and define new AI tools and workflows to enhance team productivity and system intelligence.

Nice To Haves

Previous experience working on product recommendation systems or adjacent ML-powered features (ranking, personalization, search, or similar).
Experience with big data frameworks such as Apache Spark (or similar technologies like Flink, Beam, etc.) for architecting and building complex batch or streaming pipelines.
Experience in AI/ML systems and products, such as integrating models into production systems, building features powered by ML, or contributing to the ML infrastructure.
Experience training and iterating on machine learning models (e.g., for ranking, prediction, or personalization).
Experience with ML and distributed compute frameworks such as Ray or similar tools.
Experience partnering with data science or ML teams to productionize models (designing feature stores, ensuring offline/online parity, advanced model deployment and monitoring).
Background in e-commerce, marketing tech, or consumer personalization products.

Responsibilities

Contribute to the architecture and evolution of backend services that power product recommendations across Klaviyo experiences (email, SMS, KAgent, onsite, etc.), meeting standards for reliability, performance, and clear APIs.
Contribute to and maintain robust, large-scale data processing pipelines (e.g., using Apache Spark or similar frameworks) that transform raw events and catalog data into high-quality features and inputs for recommendation models, ensuring data quality and lineage.
Collaborate closely with ML engineers and product stakeholders to productionize recommendation models —defining high-level interfaces, feature contracts, and deployment patterns for batch and/or real-time inference systems.
Contribute to the development of the vector database that powers recommendation, semantic search, and agentic use cases.
Ensure data and service observability (metrics, logging, tracing, dashboards) to facilitate recommendations that are correct, explainable, fast, and highly available for all customers.
Work with Product to break down projects into clear milestones, balancing the need for rapid experimentation with technical soundness and long-term maintainability.
Lead data-driven decision making and A/B testing efforts —ensuring recommendation systems are instrumented with the right metrics, and independently interpreting results to guide future product and engineering iterations.
Participate in on-call and incident response for the systems you own, driving major post-incident follow-ups that substantially improve the resilience and operability of our recommendation stack.
Integrate AI into your and the team’s development workflow from the ground up —for example, using AI to accelerate development, automate complex tests, or build smarter monitoring and debugging tools.
Share knowledge, mentor junior engineers, and define best practices on working with large-scale data frameworks, distributed systems, and integrating ML into production systems.