Job Title: Tech Lead / Lead Architect – RAG & Agentic AI
Location: Columbus, OH (Hybrid – 3 Days Onsite)
Duration: Long-Term Contract
Job Description
We are seeking an experienced Tech Lead / Lead Architect with deep expertise in RAG (Retrieval-Augmented Generation) and Agentic AI solutions. This role will lead the architecture, design, and delivery of enterprise-grade AI systems, working closely with customers and cross-functional teams to build scalable, secure, and high-impact AI applications.
Key Responsibilities
- Lead the architecture, design, and implementation of Agentic AI and RAG-based solutions.
- Collaborate with customers and stakeholders to gather requirements and translate them into scalable AI architectures.
- Design and build Proof of Concepts (PoCs) to validate technical approaches before production deployment.
- Develop secure, scalable, and cost-effective AI solutions on AWS.
- Implement AI governance, guardrails, observability, and evaluation frameworks.
- Drive technical discussions, mentor engineering teams, and provide architectural leadership.
- Optimize AI solutions for performance, scalability, reliability, and cost efficiency.
- Communicate architectural decisions, trade-offs, risks, and recommendations to both technical and business stakeholders.
Required Qualifications
AI & LLM Expertise
- Strong experience designing and implementing RAG pipelines.
- Hands-on expertise with:
- Embeddings
- Vector Databases
- LLM Orchestration
- Prompt Engineering Techniques
- Experience building and deploying production-grade AI/LLM solutions.
- Strong understanding of model evaluation, observability, performance tuning, and cost optimization.
- Experience implementing AI guardrails including hallucination mitigation, toxicity controls, and responsible AI practices.
AWS Expertise
Hands-on experience with: AWS Lambda,API Gateway, Amazon Bedrock, Amazon S3, Amazon OpenSearch, IAM, VPC, AWS Secrets Manager
Preferred Qualifications
- Experience with Agentic AI frameworks such as:
- Multi-Agent Systems
- MCP (Model Context Protocol) Tool Integrations
- Human-in-the-Loop Workflows