AI Engineer

Pokee AI; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

Pokee AI seeks an engineer to ship production-grade AI features at the intersection of machine learning, product, and engineering on an agentic workflow platform. You will design features powering discovery, research, and monetization, and build platforms to measure answer quality and model performance.

You will apply cutting-edge personalization, query understanding, and information retrieval techniques, and develop AI agent systems through tool design and architecture optimization.

Qualifications

• 3+ years building and shipping production AI or software systems.
• Hands-on experience with agentic applications, tool calling, and model serving.
• Strong backend and systems design skills, with practical tradeoffs between cost, latency, and quality.
• Excellent Python programming skills with production-quality code standards.
• Self-directed working style with high agency and a bias toward shipping impactful work.
• BS, MS, or PhD in Computer Science or a related field.

Responsibilities

• Design and deliver scalable AI features powering core experiences such as discovery, deep research, and monetization.
• Build evaluation platforms that measure answer quality, model performance, and downstream product impact.
• Apply state-of-the-art techniques to personalization, query understanding, and information retrieval.
• Build AI agent systems through tool design, prompt engineering, and architecture optimization.
• Continuously improve performance, reliability, and cost efficiency through thoughtful system design.

Skills

Python

Backend design

System design

Agentic apps

Shipping code

Production ML

Education

BS/MS/PhD in CS

Tools

PyTorch

TensorFlow

Model serving

RLHF

Job description

Role overview

Ship production-grade AI features at the intersection of machine learning, product, and engineering on an agentic workflow platform. The role spans feature design, evaluation infrastructure, and improvements to generative AI models and agent architectures used for discovery, research, and monetization.

Responsibilities

• Design and deliver scalable AI features powering core experiences such as discovery, deep research, and monetization.
• Build evaluation platforms that measure answer quality, model performance, and downstream product impact.
• Apply state-of-the-art techniques to personalization, query understanding, and information retrieval.
• Build AI agent systems through tool design, prompt engineering, and architecture optimization.
• Continuously improve performance, reliability, and cost efficiency through thoughtful system design.

Requirements

• 3+ years building and shipping production AI or software systems.
• Hands-on experience with agentic applications, tool calling, and model serving.
• Strong backend and systems design skills, with practical tradeoffs between cost, latency, and quality.
• Excellent Python programming skills with production-quality code standards.
• Self-directed working style with high agency and a bias toward shipping impactful work.
• BS, MS, or PhD in Computer Science or a related field.

Nice to have

• Deep expertise in PyTorch or TensorFlow, LLMs, information retrieval, NLP, or ranking systems.
• Post-training experience such as RLHF, reward modeling, SFT, or instruction tuning.
• Experience with model inference optimization and serving.
• Open-source contributions or publications in ML, NLP, or information retrieval.