Applied AI Engineer — Agentic Systems (Travel Platform)
We are building a deeply personal, AI-native travel companion designed to become the trusted travel agent for life. Travel is one of the most meaningful parts of people’s lives — yet today’s tools are fragmented, transactional, and lack continuity or understanding.
We are building something new: an intelligent, agentic system that understands users over time, applies judgment, and handles complex multi-step travel planning and booking on their behalf. The product is AI-native from the ground up, where agent behavior, reasoning, memory, and trust are core system components.
We are seeking an Applied AI Engineer to build and productionize the intelligent systems powering this agentic experience. This role focuses on translating cutting-edge ML and LLM research into reliable, scalable, user-facing systems.
What You’ll Do
Build the Core Agentic AI Layer
Develop and evolve the agentic reasoning and orchestration layer powering travel planning and booking
Design multi-step, stateful workflows for complex decision-making and execution
Implement tool-calling architectures with robust guardrails and safety constraints
Design Prompting, Memory, and Context Systems
Build prompting and orchestration strategies for reliable agent behavior
Design consent-aware long-term memory systems for personalization over time
Develop persona extraction and structured user context modeling systems
Maintain persistent, evolving representations of user preferences and intent
Build Production AI Systems
Implement autonomous booking optimization logic and decision systems
Build real-time monitoring and intervention systems for agent behavior
Ensure reliability, safety, cost-efficiency, and determinism in production environments
Evaluation and Quality Systems
Design and implement evaluation frameworks for agent performance
Measure model quality across dimensions such as:
Accuracy and reliability
Cost and latency
Safety and compliance
Determinism and consistency
Build systems for model selection, benchmarking, and continuous evaluation
Translate Research Into Production
Take state-of-the-art LLM and agent research and turn it into robust production systems
Go beyond simple LLM wrappers to build deeply integrated AI-native systems
Collaborate closely with infrastructure and product teams to ship end-to-end features
What We’re Looking For
5+ years of experience working with large-scale deep learning or LLM systems
Experience shipping production AI systems to real users at scale
Strong ability to translate research ideas into robust, reliable engineering systems
Experience building agentic workflows beyond basic LLM integration
Strong understanding of evaluation, benchmarking, and system quality measurement
Ability to reason about safety, reliability, and controllability in AI systems
Preferred Experience
Experience working with frontier LLMs (commercial or open-weight models)
Familiarity with cloud and serverless infrastructure (AWS, Vercel, etc.)
Experience with workflow orchestration systems (e.g., Temporal or similar)
Experience with LLM observability and evaluation tools (e.g., LangSmith, Braintrust, or similar)
Work on persistent memory systems and structured user representations
Experience with relational and vector databases for AI state management
Familiarity with streaming AI runtimes and tool execution frameworks
Why This Role Matters
Build the core intelligence layer of a next-generation AI-native travel agent
Define how autonomous systems reason, plan, and act on behalf of users
Work at the intersection of LLM research, systems engineering, and product impact
Shape foundational patterns for safe, reliable agentic AI systems in consumer applications
About the Company
We are a small, high-ownership team building an AI-native travel platform from first principles. We operate at the frontier of applied AI systems, focusing on reliability, evaluation, and real-world agent behavior.
We are committed to building an inclusive workplace and encourage applicants from all backgrounds to apply.