AI Enginee

Key Responsibilities
Architect RAG Systems: Design and optimize end-to-end RAG workflows using Amazon Bedrock Knowledge Bases, ensuring high retrieval accuracy and minimal hallucination.
Develop AI Agents: Build and deploy intelligent agents using Agents for Amazon Bedrock to automate multi-step tasks, integrating them with enterprise APIs and Lambda functions.
Model Selection & Tuning: Evaluate and select the best foundation models (e.g., Claude 3.5, Llama 3, Amazon Titan) for specific use cases based on performance, latency, and cost.
Vector Database Management: Implement and manage vector stores such as Amazon OpenSearch Serverless, Pinecone, or pgvector to support semantic search capabilities.
Prompt Engineering: Develop and iterate on complex system prompts and advanced prompting techniques (Chain-of-Thought, ReAct) to improve agent reasoning.
Security & Guardrails: Implement Amazon Bedrock Guardrails to ensure responsible AI practices, including PII masking and content filtering.
Performance Evaluation: Use frameworks like Ragas or TruLens to systematically evaluate RAG performance (faithfulness, relevancy) and agent success rates.


Technical Qualifications
Programming: Expert-level proficiency in Python and experience with asynchronous programming. Familiarity with Java or Node.js is a plus.
AWS Ecosystem: Hands-on experience with AWS Bedrock, Lambda, S3, DynamoDB, IAM, and Step Functions.
AI Frameworks: Deep knowledge of orchestration frameworks like LangChain, LangGraph, or LlamaIndex.
Data Engineering: Experience building ETL pipelines for unstructured data (PDFs, HTML, Markdown) to feed into knowledge bases.
DevOps/MLOps: Proficiency in CI/CD for AI, including model versioning and monitoring using Amazon CloudWatch.