Principal AI Engineer (Contract)

Overview

We are seeking a high-caliber Principal AI Engineer to accelerate the implementation of cutting-edge Agentic AI solutions. This is a hands-on builder role requiring a rare combination of deep Generative AI expertise, full-stack Python mastery, and strong AWS cloud architecture experience. You will play a critical role in transforming AI from experimental prototypes into production-grade autonomous systems that deliver real business value.

Key Responsibilities

Agentic Workflows:

• Design, build, and deploy multi-agent systems using LLM orchestration frameworks (e.g., LangGraph, CrewAI) to automate complex cross-functional business processes, with a focus on measurable efficiency gains.

Production RAG Systems:

• Develop and optimize high-performance Retrieval-Augmented Generation (RAG) pipelines using Amazon Bedrock and vector databases (e.g., OpenSearch, Pinecone), meeting defined latency and accuracy targets.

AI Integration:

• Build scalable FastAPI backends that operationalize AI model outputs.
• Collaborate with frontend teams to support React-based AI interfaces, including real-time and streaming user experiences.

Responsible AI & Guardrails:

• Implement safety mechanisms such as prompt controls, output filtering, bias evaluation, and content moderation to ensure compliance, security, and ethical AI use.

Engineering Excellence:

• Establish robust AI evaluation frameworks (e.g., Ragas, DeepEval), observability systems (e.g., LangSmith, OpenTelemetry), and CI/CD pipelines for both code and prompt lifecycle management.

Required Experience

• Software Engineering: 10+ years of professional experience
• Python Development: 7+ years of backend development using Python
• AI / Generative AI: 2+ years of hands-on experience implementing LLM-based solutions (e.g., Claude, GPT, Llama)
• AWS Cloud Architecture: 5+ years designing and deploying cloud-native applications

Technical Skills

AI & ML:

• Claude, GPT-series models, Hugging Face Transformers, PEFT, LangChain, LangGraph

Agent Systems:

• Experience with autonomous agents, tool usage, function calling, and state management

Backend Development:

• Python 3.10+, FastAPI, Pydantic, asynchronous programming

Cloud & Infrastructure:

• Amazon Bedrock, SageMaker, AWS Lambda (serverless AI), RDS, pgvector

Observability & Evaluation:

• Experience with AI monitoring, tracing, and evaluation frameworks

What Success Looks Like

• Production-ready AI agents deployed with measurable business impact
• Reliable, scalable RAG systems with strong performance benchmarks
• Secure, compliant AI systems aligned with Responsible AI standards
• Mature engineering practices applied to AI development lifecycle