ML Engineer (LATAM) - 20627

______________________________________________________________________

Role : Machine Learning Engineer (LATAM)

Location: Remote (GLOBAL)

Hours: Full-Time (US EST/Pacific Time overlap required)

Compensation: $8,000 – $11,000 USD 

About the Company: 

My client is a small, post-Series A AI company building the training data and evaluation infrastructure frontier AI labs use to improve their models, partnering with leading labs to design high-signal datasets and rigorous evaluations beyond static benchmarks. It's a lean, early team where individual contributors have direct, outsized impact on how the next generation of models learns.

Our culture blends the intellectual rigor of top quant shops with the speed and ownership of an early-stage startup experiencing hockey-stick growth.

As our first ML Specialist, you will own our recommendation and ad-serving engine end-to-end. You’ll be responsible for the real-time decision engine that determines which interactive ad reaches which user across massive concurrent traffic.

This is a full-stack ML infrastructure and engineering role (roughly 60–70% focused on systems/infra and 30–40% on modeling). You will build everything from data pipelines and feature stores to production model architecture and low-latency serving infrastructure.

What You Will Build

• Low-Latency Ad Ranking Pipeline: Architect and deploy our multi-stage recommendation pipeline from scratch (Retrieval $\rightarrow$ Ranking $\rightarrow$ Reranking).
• ML Training & Data Infrastructure: Build scalable data pipelines, feature stores, and automated model training/evaluation loops.
• Context & User Modeling: Extract high-signal embeddings and dynamic context representations from real-time conversational, engagement, and in-game signals.
• Production Serving at Scale: Design sub-second, cost-efficient, high-throughput serving systems capable of processing millions of daily requests with extreme reliability.

• Experience: 4+ years of hands-on Machine Learning Engineering experience, with at least one major production ML system shipped and maintained at scale.
• Infra + ML Hybrid Skillset: Strong backend depth spanning raw data engineering, pipeline orchestration, model deployment, and high-concurrency microservices (60–70% infrastructure focus).
• Zero-Defect Reliability Mindset: Uncompromising standards regarding latency budgets, uptime, data integrity, and cost efficiency under heavy load.
• Early-Stage Shipping Bias: High agency and fast execution pace—thriving in a 0-to-1 environment without rigid corporate playbooks.
• Communication & Location: Fluent verbal and written English; based in LATAM with reliable overlap for synchronous and asynchronous collaboration.

• Prior experience building or scaling Ad-Tech, Recommendation, or Personalization systems at scale (e.g., background at Meta, Amazon, Pinterest, TikTok, Etsy, or high-growth ad networks).
• Deep fluency with PyTorch and modern vector search/retrieval techniques.
• Advanced usage of AI-native developer workflows (Claude, Cursor, Copilot) to accelerate delivery.