Production LLM Engineer: Safe Copilots & RAG

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

Job summary

Mechabits builds copilots, RAG pipelines, and evaluation loops grounded in client data with guardrails, cost controls, and human review where it matters.

You will turn LLM ideas into production features clients can trust: retrieval that stays on their data, evaluation that catches regressions, and UX that keeps a human in the loop for high-stakes actions. We care more about shipping safe systems than flashy demos.

Qualifications

• Hands-on experience shipping LLM features beyond demos.
• Proficiency with Python and/or TypeScript service work.
• Understanding of retrieval, embeddings, and evaluation basics.
• Clear written English for async collaboration.

Responsibilities

• Design and ship RAG, copilots, and workflow automation in production products.
• Build evaluation sets, prompt versioning, and quality monitoring.
• Enforce permissions, logging, and cost controls around model usage.
• Partner with product engineers on UX that keeps humans in the loop.
• Document failure modes and safe defaults for client teams.

Skills

LLM features deployment

Python

TypeScript

Retrieval & embeddings

Evaluation basics

Job description

Mechabits builds copilots, RAG pipelines, and evaluation loops grounded in client data with guardrails, cost controls, and human review where it matters.

You will turn LLM ideas into production features clients can trust: retrieval that stays on their data, evaluation that catches regressions, and UX that keeps a human in the loop for high-stakes actions. We care more about shipping safe systems than flashy demos.