Senior AI Engineer

AI Engineer – Job Description

PURPOSE AND SCOPE

Design and deliver Generative AI solutions (LLMs, copilots, agents) that enhance automation, knowledge access, and decision-making across the enterprise.

Translate business problems into prompt-driven and retrieval-augmented (RAG) AI applications integrated with enterprise data platforms.

Ensure responsible, secure, and scalable GenAI implementations, including governance, evaluation, and monitoring of model behavior.

PRINCIPAL DUTIES AND RESPONSIBILITIES

Develop and optimize prompt engineering strategies, including prompt templates, chaining, tool use (agents), and grounding techniques (RAG).

Build and deploy LLM-powered applications using enterprise data (structured/unstructured), integrating with APIs, vector databases, and knowledge systems.

Implement evaluation, monitoring, and guardrails for GenAI systems, addressing hallucinations, bias, response quality, and performance in production.

EDUCATION

Bachelor’s or Master’s degree in Computer Science, Data Science, Engineering, or a related field.

Specialized training or certifications in AI/ML, NLP, or Generative AI preferred.

Continuous learning in emerging GenAI frameworks, LLMs, and prompt engineering techniques.

EXPERIENCE AND REQUIRED SKILLS

Hands-on experience designing and deploying Generative AI solutions on AWS and Azure, including services such as Azure OpenAI, AWS Bedrock, and Azure Machine Learning.

Strong expertise in prompt engineering, RAG architectures, embeddings, and vector databases (e.g., Databricks Vector Search, FAISS), with experience grounding LLMs in enterprise data.

Proficiency with Databricks (Spark, Delta Lake) for data preparation and integration with GenAI workflows, along with Azure DevOps and GitHub for CI/CD and GenAI application lifecycle management.

Experience with AI/LLM evaluation frameworks, including prompt/response evaluation, hallucination detection, safety filtering, and benchmark-driven optimization to ensure high-quality and reliable outputs.