Senior AI/ML Engineer - Real-Time Search & Recommendations

RiverPark Ventures; New York (NY); On-site; USD 200,000 - 280,000; Full time

Benefits offered by this job

Flexible working hours

Remote-first culture

Unlimited holiday

Growth opportunities

Job summary

Nivoda is hiring a Senior AI/ML Engineer for the Discovery Pod. You will own ranking, recommendation, and search systems end-to-end, including features, training, evaluation, deployment, and monitoring.

This role emphasizes production infrastructure, probabilistic systems, and building image-embedding search and contextual search capabilities. You will work on multi-tenant ranking, evaluation with offline metrics, and stress-testing model outputs, ensuring robust, scalable search experiences for

Qualifications

• Five to ten years of relevant experience in search and ML engineering.
• Experience building or operating real search/inference systems.
• Proficiency with NLP/LLM applications and evaluation.
• Hands-on with production ML pipelines and serving systems.

Responsibilities

• Build and operate real search infrastructure powering ranking and multi-modal search.
• Design multi-tenant ranking and recommendation systems for marketplace storefronts.
• Develop contextual search, conversational search, and image-based search pipelines.
• Own full lifecycle of ranking and recommendation models from features to monitoring.
• Design low-latency model-serving and recovery paths for probabilistic systems.

Skills

Information retrieval

Applied ML

NLP & LLM

Low latency serving

Evaluation metrics

Foundation models

Graceful failure

AI coding agents

Job description

Nivoda is hiring a Senior AI/ML Engineer for the Discovery Pod. You will own ranking, recommendation, and search systems end-to-end, including features, training, evaluation, deployment, and monitoring.

This role emphasizes production infrastructure, probabilistic systems, and building image-embedding search and contextual search capabilities. You will work on multi-tenant ranking, evaluation with offline metrics, and stress-testing model outputs, ensuring robust, scalable search experiences for