Senior AI Engineer

Reforge; Deutschland; Remote; EUR 60.000 - 85.000; Vollzeit

Zusammenfassung

A technology company is seeking an AI Engineer to enhance its platform with large language model capabilities. This fully remote role involves designing prompts, implementing LLM-powered features, and evaluating model performance. Candidates should have a strong coding background, especially in TypeScript, and excel in asynchronous communication within a distributed team.

Qualifikationen

• Experience in designing and evaluating LLMs and related products.
• Strong coding skills, particularly in TypeScript.
• Ability to work in a fully remote environment with clear communication skills.

Aufgaben

• Design and refine prompts for maximum LLM effectiveness.
• Develop and integrate new LLM-powered features.
• Evaluate LLM outputs and iterate on model settings to improve performance.
• Implement evaluation frameworks for assessing LLM performance.
• Write and maintain code for LLM interactions.

Kenntnisse

Prompt Engineering

Experience with Large Language Models

TypeScript

Python

Asynchronous Communication

Jobbeschreibung

ABOUT THIS ROLE

As an AI Engineer at Reforge, you will play a key role in shaping how our platform uses large language models to deliver value to customers. You’ll focus on building and improving LLM-powered products and features – from crafting high-quality prompts to evaluating and tuning model outputs for optimal performance. This is a highly collaborative, product-focused role where you’ll work closely with product managers and engineers to turn cutting-edge AI capabilities into practical features. This position is fully remote, so we’re looking for someone who is self-motivated and excels at clear, asynchronous communication within a distributed team.

WHO YOU'LL WORK WITH

This is a highly collaborative, product-focused role where you’ll work closely with product managers and engineers to turn cutting-edge AI capabilities into practical features. This position is fully remote, so we’re looking for someone who is self-motivated and excels at clear, asynchronous communication within a distributed team.

WHAT YOU'LL OWN

• Prompt Engineering: Design, experiment with, and refine prompts and system instructions to maximize the effectiveness and reliability of LLMs across Reforge’s products.

• LLM-Powered Features: Collaborate with product and engineering teams to develop and integrate new backend features powered by LLMs, enhancing our Insight Analytics platform with intelligent capabilities.

• Continuous Improvement: Continuously evaluate the quality of LLM outputs via internal testing and user feedback. Iterate on prompts, model settings, or data pipelines to improve performance over time and deliver a better user experience.

• Model Evaluation: Design and implement evaluation frameworks to assess LLM performance across different use cases. Monitor key metrics and develop automated testing pipelines to ensure consistent quality.

• Coding & Integration: Write and maintain code (primarily in TypeScript, with some Python) to implement support for LLM interactions – including building retrieval-augmented generation (RAG) pipelines, working with vector embeddings, and handling content chunking of large documents. A majority of your work will be in-code shipping product — not in prompting tools or notebooks.