Prompt Engineer

Prompt Engineer
Remote
May 17, 2025
Full Time
Application ends: August 17, 2025

Apply for this job

Email *

Job Description

Role Overview

We are looking for a Prompt Engineer who goes beyond clever prompt syntax. You’ll be responsible for crafting, stress-testing, and refining system prompts, user prompts, and tool-invoking instructions across a range of products built on frontier models. This is not an academic role—your work will directly affect user experience, latency, cost, and model behavior at scale.

You’ll collaborate with product designers, LLM ops, and fine-tuning teams. You must be fluent in GPT-like models, up-to-date on prompting techniques (and their tradeoffs), and able to rigorously evaluate what works—and what breaks.

Key Responsibilities

  • Architect multi-turn and context-aware prompts optimized for consistency, reliability, and performance under token constraints
  • Deconstruct product intents and translate them into concrete LLM behaviors using prompt scaffolding, formatting heuristics, and tool-calling logic
  • Prototype internal evaluations that test for prompt brittleness, hallucination risks, and boundary cases (especially when integrating plugins/tools/agents)
  • Contribute to in-house prompt libraries and interface schemas, maintaining a record of what prompt patterns generalize and which degrade over time or model versions
  • Work closely with engineers and designers to embed LLM behavior into real interfaces (e.g., smart authoring tools, summarization flows, dynamic Q&A)
  • Communicate the limitations and implicit assumptions in prompts to non-technical stakeholders

Requirements

  • Strong understanding of GPT-4/Claude/Sonar-style model behavior, especially around instruction-following, token budgeting, and role prompting
  • Demonstrated experience with prompt failure analysis: examples of what broke and how you fixed or redesigned it
  • Familiarity with system message engineering and RAG patterns (e.g., chunking strategies, retrieval-aware phrasing)
  • Able to write clear, testable prompts that operate across model versions and do not rely on non-deterministic quirks
  • Experience with prompt optimization under latency and token limits (e.g., streaming UX, batch summarization)
  • Proficient with Python or TypeScript for prototyping prompt chains and running evals (OpenAI, Anthropic, Together.ai APIs, etc.)
  • Bonus: Experience designing agent-style flows or tool-augmented completions (e.g., OpenAI Functions, LangGraph, AutoGPT-style agents)

Nice to Have

  • Past work in technical writing, support automation, tutoring platforms, or structured knowledge systems
  • Experience contributing to prompt libraries or OSS projects involving prompt tooling
  • Familiarity with model behaviors across OpenAI, Anthropic, Mistral, and open-source models (e.g., Mixtral, Phi, LLaMA)

Are you interested in this position?

Apply by clicking on the “Apply Now” button below!

#GraphicDesignJobsOnline#WebDesignRemoteJobs #FreelanceGraphicDesigner #WorkFromHomeDesignJobs #OnlineWebDesignWork #RemoteDesignOpportunities #HireGraphicDesigners #DigitalDesignCareers#Dynamicbrandguru

Related Jobs