AI Prompt Engineer

Moonfire; Ernakulam; On-site; INR 1,200,000 - 2,400,000; Full time

Job summary

heu.ai in Kochi, India, seeks a Prompt Engineer to design, test, and refine prompts that optimize the performance of AI models across its products and client solutions. This full‑time role collaborates with AI developers, data scientists, and product teams to understand project requirements and craft prompt structures and evaluation criteria.

You will run experiments to improve model accuracy, document prompt patterns and best practices, maintain a reusable prompt library, and ensure output

Qualifications

• 2+ years building software or AI/LLM components.
• Familiar with LLM fundamentals: context windows, few-shot prompting, function calling, RAG basics.
• Proficient in Python and comfortable with APIs, JSON, and Git.
• Strong documentation and written English skills.

Responsibilities

• Create and refine text prompts for AI-based video and image generation tools (e.g., OpenAI, Midjourney, Pika Labs, Seedance).
• Implement evaluation harnesses, guardrails, and regression tests so output quality is measurable and stable.
• Design, build, test, and iterate on prompts and prompt chains for production LLM workflows across client delivery and internal automation.
• Collaborate with design and marketing to produce creative visuals for campaigns.
• Experiment with different prompt styles, lighting, composition, and storytelling to achieve desired results.
• Maintain a prompt library for reuse and training.
• Ensure brand consistency and output quality across visuals.

Skills

Test and measure mindset

Written English

Documentation habits

Tools

Python

APIs

JSON

Git

Job description

Company Description

heu.ai is an AI-focused company that builds platforms and solutions to help enterprises enhance their business and tackle complex challenges. The team works closely with clients to turn business ideas into practical AI products and strategies. heu.ai supports organizations in understanding AI capabilities, defining use cases, and implementing end-to-end solutions that create real value for customers and stakeholders.

Role Description

The Prompt Engineer is responsible for designing, testing, and refining prompts that optimize the performance of AI models across heu.ai’s products and client solutions. This full-time role is based in Kochi, and involves close collaboration with AI developers, data scientists, and product teams. Day-to-day tasks include understanding project requirements, crafting prompt structures, creating evaluation criteria, and running experiments to improve model accuracy, reliability, and user experience. The role also involves documenting prompt patterns and best practices, analyzing user interactions, and iterating on prompts based on feedback and metrics. The Prompt Engineer contributes to internal knowledge sharing and supports client engagements by shaping prompt strategies aligned with business goals.

Key Responsibilities

• Create and refine text prompts for AI-based video and image generation tools (e.g., OpenAI ,Midjourney, Pika Labs, Seedance).
• Implement evaluation harnesses, guardrails, and regression tests so output quality is measurable and stays stable over time
• Design, build, test, and iterate on prompts and prompt chains for production LLM workflows across client delivery and internal automation
• Collaborate with the design and marketing teams to produce creative visuals for campaigns.
• Experiment with different prompt styles, lighting, composition, and storytelling to achieve desired results.
• Maintain a prompt library for reuse and training.
• Ensure brand consistency and output quality across visuals.

Preferred Qualifications

• Minimum 2 years building software or AI/LLM components, including hands-on experience crafting and iterating prompts for real applications
• Working knowledge of LLM fundamentals: context windows, few-shot prompting, function/tool calling, structured outputs, and RAG basics
• Proficiency in Python (or similar) and comfort with APIs, JSON, and version control (Git)
• A test-and-measure mindset — you evaluate output quality, you don't eyeball it
• Clear written English and strong documentation habits