LLM Applications Engineer

SupportFinity™; New York (NY); Hybrid; USD 130,000 - 175,000; Full time

Job summary

SupportFinity™ is seeking an LLM Applications Engineer located in New York to build and deploy production-grade LLM-powered systems. This hybrid role involves hands-on work across backend Python systems and React-based frontend applications, focusing on designing RAG pipelines and orchestration layers.

The ideal candidate will have 2+ years of full-stack experience, strong proficiency in Python and JavaScript, and a passion for deploying LLM systems in production environments. The role uniquely intersects AI infrastructure and product development.

Qualifications

• 2+ years of full-stack web development experience.
• Hands-on experience building production RAG pipelines and retrieval systems.
• Ability to operate across backend infrastructure and frontend UI.

Responsibilities

• Design and deploy production-grade RAG pipelines.
• Build and optimize vector retrieval systems.
• Collaborate closely with product and engineering teams.

Skills

Full-stack web development experience

Proficiency in Python

Proficiency in JavaScript or TypeScript

Experience with RAG pipelines

Strong API design fundamentals

Experience with SQL databases

Education

Computer Science degree from a top-tier program

Tools

LangChain

LlamaIndex

Haystack

Job description

Location: New York | San Francisco | Munich | London (In-Person)

Employment Type: Full-Time

Base Salary: $130,000 – $175,000

Overview

We are hiring an LLM Applications Engineer to build and deploy production‑grade LLM‑powered systems. This is a hybrid AI infrastructure and product engineering role. You will design and implement RAG pipelines, vector retrieval systems, agentic workflows, and full‑stack LLM‑powered product experiences. The role requires hands‑on ownership across backend Python systems and React‑based frontend applications. This position is for engineers who move beyond prototypes and ship robust, scalable LLM systems into production.

What You’ll Do

• Design and deploy production‑grade RAG (Retrieval‑Augmented Generation) pipelines
• Build and optimize vector retrieval systems
• Implement agentic LLM workflows and orchestration layers
• Develop full‑stack product experiences powered by LLMs
• Design clean, scalable APIs and asynchronous processing systems
• Connect LLM systems to structured data sources, including SQL databases and data engines
• Collaborate closely with product and engineering teams to ship LLM‑first features

What We’re Looking For

Core Requirements

• 2+ years of full‑stack web development experience
• Proficiency in Python and JavaScript or TypeScript
• Experience with LLM orchestration frameworks such as LangChain, LlamaIndex, or Haystack
• Hands‑on experience building production RAG pipelines and retrieval systems
• Strong API design and asynchronous processing fundamentalsAbility to operate across backend infrastructure and frontend UI
• Computer Science degree from a top‑tier program

Strong Plus

• Built and deployed RAG pipelines in live production environments
• Experience working with scientific, technical, or research datasets
• Strong product mindset with LLM‑first feature development
• Experience integrating LLM systems with SQL databases and broader data infrastructure

Who This Is Not For

• Have only academic or prototype‑level LLM exposure
• Have exclusively backend‑only or frontend‑only experience
• Lack experience with vector databases or retrieval systems
• Have not deployed LLM systems into production environments

What Success Looks Like

• Production‑grade RAG pipelines running reliably at scaleClean, well‑architected retrieval and orchestration systems
• Seamless integration between LLM backends and frontend product experiences
• Measurable product impact from LLM‑powered features
• Ownership of end‑to‑end LLM application architecture

Why This Role Is Unique

This is not a research‑only AI role and not a traditional full‑stack position. You will sit at the intersection of AI infrastructure and product, building systems that combine retrieval, orchestration, and real‑world user interfaces. If you want to ship meaningful LLM‑powered products — not just experiment with models — this is an opportunity to own that architecture end to end.