Note: The job is a remote job and is open to candidates in USA. Beacon is an early-stage, high-growth technology startup building a data-intensive AI platform designed to transform how complex product and commerce data is researched, structured, and operationalized. The company is seeking a Product Engineer who can build production systems around messy, high-volume data and emerging AI capabilities, owning systems from initial experimentation through production deployment and iteration.
Responsibilities
- Build ingestion, normalization, and enrichment pipelines processing millions of records
- Develop search, retrieval, ranking, RAG, structured extraction, and entity resolution systems
- Build long-running AI workflows and research agents that can execute across hundreds of model calls and recover reliably from failures
- Design production APIs, backend services, and customer-facing integrations
- Work with distributed systems concepts including queues, workers, retries, idempotency, and eventual consistency
- Build evaluation frameworks, golden datasets, tracing, and metrics to understand whether AI and retrieval systems are actually improving
- Improve the reliability, observability, performance, and cost of production AI systems
- Own problems end-to-end, from shaping the idea and building an MVP to testing in production and iterating based on results
- Work closely with customers to understand complex data and workflow problems
Skills
- Strong backend and ML engineering experience building production systems
- Experience working with large datasets and designing reliable data pipelines or distributed systems
- Experience shipping search, recommendation, ranking, information retrieval, entity resolution, extraction, or RAG systems to production
- Strong understanding of how to evaluate system quality using metrics such as precision, recall, ranking metrics, offline evaluation, online experiments, A/B testing, or golden datasets
- Experience experimenting with and iterating on AI, ML, search, or retrieval systems based on measurable results
- Ability to build reliable production infrastructure around probabilistic or non-deterministic AI systems
- Strong experience with APIs, databases, and backend system design
- Experience with TypeScript and modern backend development
- Experience working with Postgres and large-scale analytical data systems such as BigQuery
- Willingness to work directly with customers and deeply understand their workflows
- Experience building AI agents or long-running agentic workflows
- Experience with LLM evaluation, tracing, and observability
- Experience with Temporal or similar durable workflow orchestration systems
- Experience building data products involving millions of records or entities
- Experience with product catalogs, commerce data, marketplaces, or complex entity matching
- Experience at an early-stage or high-growth startup
Company Overview