Senior Full Stack Engineer

TourHero

7h•Remote

About The Position

TourHero is a social travel platform that enables anyone to meet like-minded individuals through unforgettable journeys, bringing us closer as humans, one journey at a time . We're building an online marketplace to help plan unique epic adventures that enables travellers to meet other like-minded new people during their trips. We are an early stage start-up, backed by the 2 of the largest and most iconic venture capitalists in the world. We are building a strong team to execute this grand vision. We’re looking for someone different - someone with an entrepreneurial heart. We’re looking for someone who’ll be willing to get their hands dirty in touching every part of the business and pushing the boundaries. Are you that person? If so, this is the perfect match for you! As a Senior Full Stack Engineer (AI Product Focus) at TourHero, you’ll be the driving force behind our mission to create unforgettable travel experiences. Leveraging your expertise across frontend, backend and AI-powered product development, you’ll craft user-centric solutions that redefine the way people explore the world. You’ll work directly with the founding team, product and design teams, partnering with skilled colleagues to build robust systems, captivating interfaces, and AI experiences (LLM-powered) that inspire and delight users throughout their journey.

Requirements

5+ years of experience in a full stack engineering role or similar capacity.
Demonstrated expertise in at least one modern frontend framework (e.g React, Angular, Vue.js).
Proficiency in HTML, CSS, and JavaScript, with a strong grasp of web standards and accessibility principles.
Experienced with databases and proficient in data modelling.
Hands-on experience building AI/LLM-powered products in production, including prompt engineering and iteration.
Strong understanding of prompting techniques (few-shot, chain-of-thought style decomposition, structured outputs, system vs user instructions, context management).
Practical knowledge of embeddings, vector search, and RAG (chunking, retrieval strategies, re-ranking, grounding).
Experience integrating with LLM APIs and building application-level guardrails (validation, moderation, policy constraints).
Familiar with evaluation and monitoring for AI features (quality metrics, regression testing for prompts, A/B tests, feedback loops).
Knowledgeable in software engineering best practices, including version control, testing methodologies, and CI/CD.
Strong problem solving skills and meticulous attention to detail.
Familiarity with Agile development methodologies.
Exceptional communication and collaboration abilities.
Thrives in the energetic pace of startup environments.
Passionate about travel; excited about our mission to create unforgettable experiences for our customers.

Nice To Haves

Proficiency in Ruby on Rails
Experience with fine-tuning / adapters, or managing datasets for supervised preference tuning.
Experience with LLM orchestration frameworks (optional) and/or building without them.
Familiarity with model risk & safety practices (prompt injection mitigation, data leakage prevention).
Experience with recommendation systems or matching algorithms (even lightweight).

Responsibilities

Collaborate closely with product managers and designers to conceptualise, design, and implement new features, ensuring alignment with user needs and business objectives.
Write efficient clean and reusable code that can be easily maintained and scaled.
Design and build AI-powered product features (e.g., itinerary generation, smart recommendations, chat-based trip planning, community matching) that feel magical and reliable.
Own prompt design and prompt iteration: create, test, version, and continuously improve prompts to deliver consistent, high-quality outputs aligned with our brand voice and product goals.
Implement LLM application patterns such as: Retrieval-Augmented Generation (RAG) using embeddings + vector databases Tool/function calling and lightweight “agentic” workflows Structured output generation (schemas/JSON), validation, and guardrails
Build systems for evaluation and quality: automated test suites for prompts, offline evaluation sets, human-in-the-loop review flows, and output scoring.
Optimize for latency, cost, and reliability: caching, streaming responses, fallbacks, rate-limit handling, and monitoring.
Partner with product/design to craft UX for AI (confidence, citations where relevant, editable outputs, “regenerate” flows, and user controls).
Ensure privacy, safety, and compliance in AI features (PII handling, data retention, content filtering, and abuse prevention).