Remote AI Engineer - Production LLM & Agentic Systems

datadotorg; Union City (NJ); A distancia; USD 105.000 - 125.000; Jornada completa

Descripción de la vacante

datadotorg in Union City, NJ (remote) seeks a mid-level ML engineer to own AI features from day one. You will work with the VP of Engineering on system design, guide architectural choices, and deliver production AI capabilities with strong collaboration.

The role emphasizes prompt engineering, integration work, and safety guardrails, with hands-on coding, reviews, and ongoing performance tuning as you scale AI across our data products.

Formación

• 2–3+ years building or deploying ML models or AI-driven apps.
• Proficient in Python with ML fundamentals and API integration.
• Hands-on experience with foundation models and agentic patterns.

Responsabilidades

• Build and ship LLM-powered features with prompt orchestration and tool calls.
• Develop API integrations and RAG workflows connecting data systems to vector search.
• Create evaluation infrastructure and LLM-as-a-judge assessments.
• Monitor production AI features for accuracy, latency, and costs; tune prompts.
• Implement runtime safety and reliability guardrails within compliance bounds.
• Research orchestration frameworks and model providers; advise on build-vs-buy decisions.
• Collaborate with non-technical stakeholders to scope AI problems and deliver solutions.
• Evaluate emerging models/tools and contribute to code reviews to raise standards.

Conocimientos

Python

ML fundamentals

LLM systems

APIs

Prompt engineering

RAG

LangChain

Vector stores

Educación

Bachelor's in CS/Engineering/AI/ML

Herramientas

LangChain

LlamaIndex

pgvector/OpenSearch

Descripción del empleo

datadotorg in Union City, NJ (remote) seeks a mid-level ML engineer to own AI features from day one. You will work with the VP of Engineering on system design, guide architectural choices, and deliver production AI capabilities with strong collaboration.

The role emphasizes prompt engineering, integration work, and safety guardrails, with hands-on coding, reviews, and ongoing performance tuning as you scale AI across our data products.