Senior AI Engineer

Teak is reinventing refunds with a proprietary SaaS embedded refund protection platform. The role will help build an agentic layer on top of existing platform APIs for compliant, AI-native refund solutions.

Responsibilities

• Architect and build the agentic layer end-to-end
• Design, build, and maintain production-grade agentic systems
• Develop MCP servers and tool integrations to connect AI agents to Teak and partner systems
• Orchestrate LLMs across the AI layer and manage prompt and system-prompt strategies
• Build RAG pipelines grounded in Teak policy and product data
• Implement guardrails for compliant offer language and regulatory requirements
• Build evaluation frameworks to monitor and improve agent performance and safety
• Build and operate AI infrastructure on AWS with auditability and rapid issue resolution
• Contribute to backend services, APIs, and platform improvements in Python
• Participate in code reviews, document system design and integration patterns, and join Agile workflows
• Maintain production-grade systems beyond prompt work through design, deployment, and operation
• Stay closely connected through code reviews, technical discussions, and regular team syncs
• Apply clean code, testing, and strong engineering fundamentals
• Join the on-call rotation

Requirements

• 5+ years of professional software engineering experience
• At least 2 years focused on production AI/LLM application development
• Strong Python proficiency
• Deep backend engineering experience with production services and APIs
• Working knowledge of Model Context Protocol (MCP) and tool-calling patterns
• Experience designing and running LLM evaluation frameworks for quality, reliability, and safety
• Experience building and operating AI infrastructure on a major cloud platform
• Reliable internet connection and appropriate home office workspace required

Qualifications

• Bachelor's Degree in Computer Science, Engineering, a related field, or equivalent practical experience

Skills

• LLM
• RAG
• LangChain
• Anthropic
• OpenAI
• LangGraph
• MCP
• AWS Bedrock
• Machine Learning
• Python
• Django
• Vector Search
• AWS
• GCP
• Azure
• Agentic AI systems
• Multi-agent orchestration
• Human-in-the-Loop workflows
• Context management
• Memory
• Tool-calling
• Claude, GPT, Gemini
• Knowledge systems
• Policy data
• Product data
• Guardrails
• Compliance
• Structured logging
• Tracing
• APIs
• Testing
• Code quality
• System design
• Agile workflows
• On-call rotation

Nice To Have

• Background in a compliance-sensitive industry such as insurance, fintech, or legal
• Experience operating LLM systems at high volume
• Attention to latency, cost, and caching
• Familiarity with AWS Bedrock or similar managed AI platforms and agent runtimes
• Experience with model fine-tuning, customization, or distillation
• Contributions to open-source AI tooling or frameworks

Pay

$36,100–$48,200/yr

Schedule

Full-time, remote