AI Engineer

Role: AI Engineer

Location: Plano, TX

Go, Python, Java, Terraform, C2, S3, Terraform/CloudFormation, Lambda, DynamoDB "Troubleshooting Skills - 5 years:

Systematic debugging

Log analysis

Performance profiling

Incident response

Prompt Engineering - 2 years:

Prompt Design: effective prompts for different personas, tasks, schema & models Few-shot & Zero-shot Learning: guide model behaviors Tool use integration: prompt to interact with APIs, Databases and external tools Building AI Agents - 2 years:

Agent Frameworks: langchain, autogen, crewai, MCP, etc.

Memory & Context Management: long-term memory, RAG Tool Use & Planning: Integrating external tools (API/DB), multi-step reasoning, task decomposition State management: goal tracking, actions, outcomes LLM Integration: using common open and closed models (OpenAI, Gemini, Anthropic, Llama, etc.) Evaluation & safety: testing agent behavior, fail-safes, ethical constraints

Security: input sanitization, sensitive data handling, prompt injection projection, scoped permissions, logging, rate limiting"

Bedrock, Bedrock Agents, IAM policies, roles, KMS, secrets manager

Role Descriptions: "Troubleshooting Skills 5 years:Systematic debuggingLog analysisPerformance profilingIncident responsePrompt Engineering 2 years:Prompt Design: effective prompts for different personas| tasks| schema & modelsFew-shot & Zero-shot Learning: guide model behaviorsTool use integration: prompt to interact with APIs| Databases and external toolsBuilding AI Agents 2 years:Agent Frameworks: langchain| autogen| crewai| MCP| etc.Memory & Context Management: long-term memory| RAGTool Use & Planning: Integrating external tools (API/DB)| multi-step reasoning| task decompositionState management: goal tracking| actions| outcomesLLM Integration: using common open and closed models (OpenAI| Gemini| Anthropic| Llama| etc.)Evaluation & safety: testing agent behavior| fail-safes| ethical constraintsSecurity: input sanitization| sensitive data handling| prompt injection projection| scoped permissions| logging| rate limiting"