Senior AI/ML Engineer

This is a fully remote position, open to applicants in New York, +1 more state.

📋 Description

• Develop and maintain search infrastructure to enhance ranking and multi-modal search capabilities across Nivoda’s marketplace.

• Create multi-tenant, configurable ranking and recommendation systems for both Nivoda’s marketplace and personalized jeweller storefronts.

• Enhance contextual search and a forthcoming conversational assistant through prompt design and retrieval-augmented methods.

• Construct pipelines for image-to-attribute and image-to-product matching utilizing image embeddings.

• Design recovery pathways and confirmation steps for probabilistic systems prior to high-stakes actions.

• Define and conduct search- and ranking-specific evaluations, incorporating offline ranking metrics and interleaving experiments.

• Assess and stress-test outputs from foundation models and LLMs.

• Manage the complete lifecycle of ranking and recommendation models, from feature selection and training to evaluation, deployment, and monitoring.

• Design and optimize low-latency, real-time model-serving systems under production conditions.

• Leverage AI coding agents to develop ML infrastructure and pipelines with stringent evaluation frameworks.

• Collaborate within the Discovery Pod to build ranking, recommendations, search input, contextual search, conversational search, and image-upload search for a global jewellery marketplace.

⛳️ Requirements

• 5–10 years of relevant experience.

• Experience with information retrieval infrastructure utilizing Elasticsearch/OpenSearch, Vespa, or vector-search databases like pgvector, Pinecone, or Weaviate.

• Practical machine learning experience for ranking and recommendations, encompassing learning-to-rank, embedding-based retrieval, and collaborative filtering.

• Proven track record owning features, training, evaluation, deployment, and monitoring of ML models.

• Background in production NLP and LLM applications, including prompt design, retrieval-augmented techniques, and managing zero-result and tradeoff situations.

• Experience with low-latency serving in real-time model-serving systems under production load.

• Skills in search and ranking evaluation, including offline ranking metrics and interleaving experiments.

• Hands-on experience with foundation models and LLMs, particularly in evaluating and stress-testing model outputs.

• Experience in designing effective failure handling for probabilistic systems.

• Proficiency in utilizing AI coding agents to construct ML infrastructure and pipelines, with a strong focus on evaluation and testing discipline.

• Curriculum vitae in English.

• Nice to have: fundamentals of computer vision and experience with image embeddings.

• Nice to have: experience in designing multi-tenant configurable systems.

• Nice to have: experience with conversational or assistant-style search products.

• Nice to have: background in marketplace or e-commerce search, ranking, or recommendations.

🏝️ Benefits

• Flexible working hours.

• Remote-first culture.

• Genuine opportunities for growth and learning.

• Unlimited holiday allowance.

• Chance to join during a phase of exponential expansion.