SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Nivoda is a digital marketplace operating system for the global jewellery industry, connecting diamond and gemstone suppliers with retailers across 70+ countries. The company powers over $300M in annual transactions within a $350B+ industry and is early in digitizing one of the last major offline global markets.
You'll join the Discovery Pod, a small, high-ownership team building AI-powered search, ranking, and recommendation systems that help hundreds of thousands of diamonds and gemstones get discovered. This is production-grade information retrieval and applied ML work—not API wrappers or hosted search widgets. You'll own the full lifecycle of ranking and recommendation models, from feature engineering through training, evaluation, deployment, and monitoring.
Core responsibilities include:
• Build and operate real search infrastructure (OpenSearch, Vespa, or vector-search databases) powering ranking and multi-modal search across Nivoda's marketplace and individually branded jeweller storefronts.
• Design multi-tenant, configurable ranking and recommendation systems that serve both Nivoda's marketplace and many independently branded storefronts without code forking.
• Build contextual search and a planned conversational assistant using prompt design and retrieval-augmented generation, handling zero-result and tradeoff scenarios gracefully rather than returning confidently wrong answers.
• Build image-to-attribute and image-to-product-match pipelines using image embeddings for photo-based search.
• Define and run search-specific evaluation—offline ranking metrics, interleaving experiments—to determine whether ranking changes are genuinely better.
• Own the full ML lifecycle for ranking and recommendation models using learning-to-rank, embedding-based retrieval, and collaborative filtering techniques.
• Design and optimize low-latency, real-time model-serving systems that keep search instant under production load.
• Use AI coding agents to build ML infrastructure and pipelines, paired with rigorous evaluation harnesses that catch ranking and retrieval bugs.
Tech stack: OpenSearch (fed via CDC from PostgreSQL), Spark/EMR/Glue ingestion, PostgreSQL (OLTP) → Snowflake (dbt-modeled medallion layers), TypeScript/Node.js backend, GraphQL API gateway, Nx monorepo, Jenkins CI/CD, AWS (EMR, Glue, DMS, Kubernetes), Grafana observability.
Requirements:
• 5–10 years of experience
• Information retrieval infrastructure experience—building or operating real search systems (Elasticsearch/OpenSearch, Vespa, pgvector, Pinecone, Weaviate)
• Applied ML for ranking and recommendations end-to-end (learning-to-rank, embedding-based retrieval, collaborative filtering) with hands-on ownership of features, training, evaluation, deployment, and monitoring
• NLP and LLM application experience—production experience building on language models, including prompt design, retrieval-augmented approaches, and graceful handling of zero-result and tradeoff scenarios
• Low-latency serving experience—built and optimized real-time model-serving systems under production load
• Specialized evaluation skills for search and ranking quality (offline ranking metrics, interleaving experiments, ability to judge whether a ranking change is genuinely better)
• Direct, hands-on experience building with foundation models and LLMs as engineering material—prompt design, retrieval-augmented approaches, and critically, evaluating and stress-testing model output rather than trusting it by default
• Experience designing for graceful failure in probabilistic systems—knowing when a model is confidently wrong and building product around that reality
• Comfort using AI coding agents to build ML infrastructure and pipelines with strong evaluation and testing discipline
Nice to have:
• Computer vision fundamentals, specifically image embeddings
• Multi-tenant, configurable system design experience
• Conversational or assistant-style search product experience
• Background in marketplace or e-commerce search, ranking, or recommendations
