Machine Learning Engineer - Recommendations (AI)

Location: Remote - Americas

Company

hirify.global provides commerce infrastructure that helps entrepreneurs and enterprises build and scale businesses, serving millions of merchants in 175 countries.

What you will do

• Develop and optimize machine learning models for recommendation retrieval, personalization, and multi-objective ranking.
• Build and maintain end-to-end ML pipelines covering feature engineering, training, offline evaluation, online experimentation, and production serving.
• Scale LLM-based recommendation systems, including throughput optimization, QPS management, multi-partition workers, and operational reliability.
• Collaborate with platform, infrastructure, and product teams to integrate ML solutions into the unified recommendations pipeline.
• Research and implement semantic and sequence models, distillation approaches, and relevance grading to improve recommendation quality.
• Document technical insights, share best practices, and contribute to experimentation practices across engineering teams.

Requirements

• Extensive experience building and deploying machine learning models for large-scale recommendation, personalization, or ranking systems.
• Strong proficiency with TensorFlow or PyTorch and expert-level Python programming skills.
• Hands-on experience with learning-to-rank methods and gradient-boosted tree frameworks including XGBoost, CatBoost, or LightGBM.
• Experience processing and analyzing large-scale datasets, with familiarity with BigQuery, dbt, feature stores, and streaming event pipelines.
• Experience with statistical methods and performance optimization for ML systems at scale.
• Exposure to MLOps and control planes, including orchestration, infrastructure-as-code, and runtime hardening.
• Experience with production LLM systems, prompt or parameter optimization, throughput scaling, semantic embeddings, two-tower architectures, or streaming serving such as Flink.

Culture & Benefits

• Digital-first daily work in a fast-moving and change-oriented environment.
• Strong focus on experimentation, technical knowledge sharing, and continuous improvement.
• The role may require on-call work.
• The hiring process is designed to move quickly, with the interview loop targeted for completion within 30 days.

Hiring process

• Candidates should be prepared to interview within one week of applying.
• The process includes a live pair-programming session; candidates should bring their own IDE.
• AI-enabled tools may be used to screen, select, and assess applications, with recruitment-team review of outputs.