Prompt Engineer (Mid-Level) (Spain based)

Hubtype is a platform that helps companies deploy and scale AI agents across messaging channels like WhatsApp. These agents help customers get things done, tracking orders, managing bookings, asking questions, resolving issues, and receiving recommendations — all in a single conversation.

Today, millions of conversations run through Hubtype every month for brands such as EasyJet, Wizz Air, JetSMART, Securitas Direct, Occident, and Retorna. Customer expectations have shifted toward experiences that feel fast, personal, and effortless, and messaging is where those experiences now happen.

We are in a particularly exciting phase: moving from structured automation into AI agents that can self-solve, reason, and adapt — while still giving enterprises the control and reliability they need. It’s not a simple problem. That’s exactly why it matters.

🎯 Purpose of the Role

We are looking for a Prompt Engineer with a strong technical foundation and a creative, experimental mindset.

This role sits at the intersection of artificial intelligence, product design, and applied research.

You’ll work closely with Engineers to design, optimize, and evaluate prompts for large language models (LLMs), multimodal systems, and embedding-based architectures.

The ideal candidate is curious, detail-oriented, and passionate about translating complex business needs into effective prompt strategies that enhance performance and reliability in generative AI applications.

Responsibilities

• Design, iterate, and optimize prompts for generative models, including LLMs, multimodal models, and embedding-based systems.
  
  
• Apply advanced prompting techniques such as few-shot, zero-shot, prompt chaining, and agentic prompting.
  
  
• Document and systematize prompt patterns, ensuring consistency, traceability, and scalability across use cases.
  
  
• Collaborate closely with Product Engineers and Product Managers to prepare prompts and prompt chains for integration into agent architectures and Retrieval-Augmented Generation (RAG) systems.
  
  
• Define and implement EVALs (evaluation frameworks) to assess model performance, accuracy, and alignment.
  
  
• Translate business requirements and user needs into effective and structured prompt designs.
  
  
• Contribute to the continuous improvement of internal methodologies and knowledge bases for prompt design.

• Participate in experimental and prototyping initiatives, testing new prompting strategies and evaluation methods.

Requirements

• Proven experience in prompt design and optimization for LLMs, multimodal, or embedding-based systems.
  
  
• Deep understanding of advanced prompting techniques (prompt chaining, few-shot, zero-shot, agentic prompting).
  
  
• Strong ability to document, version, and systematize prompt workflows.
  
  
• Familiarity with evaluation metrics and quality assessment frameworks for generative model outputs.
  
  
• Ability to collaborate cross-functionally with technical and non-technical teams.
  
  
• Creative and analytical mindset, with a focus on experimentation and continuous improvement.
  
  
• Solid technical background (e.g., in Computer Science, AI, Data Science, or related fields).
  
  
• Minimum 2 years of experience in Software Engineering or a similar technical role.

Benefits

• A role where you'll have lots of room to shine
• Working with amazing people from all over the world, with different cultures, experiences and knowledge
• 29 days of holiday per year, plus national holidays
• Fully remote policy and very flexible working – we trust you to get the work done
• 10% personal growth time. At Hubtype you’re expected to dedicate 10% of your time to self-improvement in the way you think it’s best for you. Whether it’s through online courses, side projects or reading books, that’s up to you.
• Attractive Benefits & Perks Plan, centered in Learning&Development
• Workstation budget for make great work happen at home

Want to See Our Impact?

See Hubtype in Action

• The future of sales lives inside WhatsApp

Hear from some of our customers:

• JetSMART reduced 34% call center volume and automated 80%+ of customer interactions https://www.hubtype.com/success-stories/jetsmart
• easyJet increased Trustpilot score by 107% by automating feedback journeys https://www.hubtype.com/success-stories/easyjet
• Securitas Direct improved successful lead engagement by shifting to WhatsApp https://www.hubtype.com/success-stories/securitas-direct

Read these explanatory blog posts

• WhatsApp is no longer just a Channel. It’s a Business Platform: https://www.hubtype.com/blog/whatsapp-as-a-business-platform
• Creating an Inclusive Culture at Hubtype https://www.hubtype.com/blog/inclusive-culture-hubtype
• Curious about the vibe?. Here’s what our last offsite looked like

Got questions? Feel free to get in touch.

If you're ready to apply, here's what we'd like to see:

• An up-to-date CV.
• A motivation letter about yourself and why you think you fit the position. Be as specific as you can.

If you want to read more, check us out at https://www.hubtype.com/. We look forward to hearing from you! 👋