Tech Lead Lead Architect – RAG & Agentic AI
Job Location:
Wilmington, DE - USA
Monthly Salary:
Not Disclosed
Posted on:
7 days ago
Vacancies:
1 Vacancy
Job Summary
Tech Lead / Lead Architect RAG & Agentic AI
Role Summary:
Lead architecture design and delivery of Agentic AI and RAG-based solutions partnering with customers and internal teams to build scalable secure and high-impact AI systems.
Must-Have:
- Strong experience in RAG pipelines embeddings vector DBs LLM orchestration and prompting techniques.
- Hands-on expertise in AWS (Lambda API Gateway Bedrock S3 OpenSearch IAM VPC Secrets Manager).
- Ability to design end-to-end AI architecture and build PoCs before committing solutions to customers.
- Deep understanding of AI guardrails (toxicity hallucination control) data privacy and cloud security patterns.
- Proven ability to lead from the front mentor teams and own delivery under tight timelines and high visibility.
- Strong customer communication skills ability to explain architecture trade-offs and risks clearly.
- Experience handling model evaluation observability performance tuning and cost optimization in production AI systems.
- Expertise in API design microservices integration and event-driven architectures for AI systems.
Good-to-Have:
- Experience with Agentic AI frameworks (LangGraph CrewAI AutoGen Semantic Kernel etc.).
- Exposure to marketing domain use cases (campaign optimization personalization analytics insights).
- Familiarity with multi-agent orchestration tool usage (MCP) and human-in-loop workflows.
Technical Fit
- Can clearly explain a RAG architecture (data ingestion embedding retrieval generation)
- Has built or deployed production AI/LLM solutions (not just POCs)
- Understands agent lifecycle orchestration and tool integrations
- Demonstrates AWS architecture security (IAM roles network isolation secrets)
- Knows prompt engineering evaluation guardrails implementation
Architect & Leadership Fit
- Has led architecture/design discussions with customers
- Can drive PoC production transition independently
- Shows ownership mindset (decision-making without dependency)
- Can mentor/coach developers and review designs/code