Role : AI Prompt Engineer

Relevant Experience : 3 years

Location : Bangalore

Work mode : 5 days from office

Notice period : 30 days

Role Summary :

The AI / Prompt Engineer is responsible for designing, configuring, and continuously tuning generative AI prompts and retrieval workflows that power Aezions AI-driven QA scoring, coaching intelligence, and knowledge retrieval capabilities. This role ensures that AI outputs are accurate, explainable, compliant, and aligned with Client-defined quality frameworks, including healthcare regulatory requirements.

Key Responsibilities :

Prompt & AI Logic Design :

- Design, develop, and maintain prompt templates for :

1. AI-assisted QA scoring

2. Compliance detection and risk flagging

3. Interaction summaries and evidence extraction

- GenAI-powered coaching recommendations and session summaries

- Implement multi-step prompt chains (reasoning scoring evidence recommendation).

- Enforce guardrails to minimize hallucinations and ensure grounded outputs via RAG.

RAG & Knowledge Grounding :

- Define chunking, embedding, and retrieval strategies for structured and unstructured content.

- Ensure AI responses are strictly grounded in approved source documents and interaction data.

- Tune retrieval thresholds, ranking logic, and context window usage for accuracy and consistency.

- Validate citation and evidence generation for audit and compliance purposes.

Healthcare & Compliance Alignment :

- Translate QA scorecards, policies, and compliance rules into AI-interpretable logic.

- Collaborate with healthcare SMEs to encode regulatory requirements into prompts and validation rules.

- Ensure explainability of AI outputs for auditability and regulatory review.

Model Configuration & Tuning :

- Configure and tune LLM usage across supported platforms (e.g., Amazon Bedrock).

- Optimize prompt performance for accuracy, latency, and cost efficiency.

- Support model versioning, prompt lifecycle management, and rollback strategies.

Testing, Calibration & Quality Control :

- Develop test cases and evaluation datasets for AI outputs.

- Participate in AI vs. human QA calibration sessions.

- Monitor prompt drift, performance degradation, and unintended behavior.

- Iterate prompts based on feedback, edge cases, and evolving requirements.

Collaboration & Documentation :

- Work closely with Product Owner, Business Analyst, and Data Engineering teams.

- Document prompt logic, assumptions, and limitations.

- Contribute to Requirements Specifications and AI design artifacts.

- Support knowledge transfer and operational readiness.

Required Skills & Experience :

Technical Skills :

- Hands-on experience with large language models (LLMs) and prompt engineering.

- Experience implementing Retrieval-Augmented Generation (RAG) architectures.

- Familiarity with AWS AI services (Amazon Bedrock, Comprehend, Transcribe preferred).

- Understanding of embeddings, vector search, and context management.

- Ability to structure AI outputs into consistent, machine-consumable formats (e.g., JSON).

Domain & Analytical Skills :

- Strong analytical mindset with attention to detail.

- Experience translating business rules or scorecards into system logic.

- Exposure to regulated environments (healthcare, financial services, insurance preferred).

- Understanding of QA processes, contact center operations, or coaching workflows is a plus.