F
Artificial Intelligence Engineer
FORT
3 hours ago
Full-time
On-site
New York City, New York, United States
Applied AI Engineer – Conversational Intelligence - 5 days a week. xz
The Opportunity
We’re partnering with a well-funded, early-stage consumer AI company building an agentic platform that turns natural-language requests into real-world actions.
This is not a chatbot or experimental AI feature. The product combines conversational intelligence, live inventory and personalised recommendations to help consumers discover and book experiences through a single conversation.
You’ll join a small, exceptionally fast-moving engineering team and take end-to-end ownership of the product’s conversational intelligence layer. Your job is to make the system consistently produce accurate, useful actions—not simply convincing responses.
What You’ll Own
The complete agent architecture, including prompt systems, tool calling, multi-step reasoning and streaming
Conversational state, memory and preference-recall infrastructure
LLM-powered ranking, retrieval and personalised recommendations
Multi-turn reliability, including refinement loops, error recovery and graceful degradation
The rendering layer between agent outputs and the consumer-facing product
Model-provider strategy across platforms such as OpenAI, Anthropic and Gemini
Structured outputs, caching, routing and fallback behaviour
Evaluation infrastructure for agent, retrieval and recommendation quality
Investigation and resolution of real-world production failure modes
What We’re Looking For
3+ years of experience as a backend or full-stack software engineer
Experience building and shipping consumer-facing products—not solely internal tools or B2B software
Proven experience taking LLM-powered ranking, retrieval, search or recommendation systems into production
Experience building systems in which a model interprets user preferences and surfaces relevant results
Broad capability across the applied AI stack, including prompting, embeddings, retrieval, memory and personalisation
Practical experience defining, running and improving evaluations for ranking or retrieval quality
The ability to explain specific AI-system failure modes you have diagnosed in production
Strong product judgement and a focus on user outcomes
Regular use of AI-native tools and workflows in your own development process
Technology The primary application stack is TypeScript, although strong Python engineers are also encouraged to apply. Direct experience with every tool is not required; we care more about engineering fundamentals, learning speed and evidence that you can ship reliable AI products.
You’ll Thrive Here If You
Move quickly without lowering your quality bar
Enjoy owning ambiguous, technically difficult product problems
Think across the entire system rather than specialising in one narrow part of the AI stack
Test ideas by building and putting them in front of users
Care deeply about reliability, latency and the quality of the final experience
Have side projects or other evidence of independent, AI-native building
Why Join?
Own a critical product layer from architecture through production
Work on a genuinely agentic consumer experience with real-world utility
Join a small team with unusually high individual ownership
Build with strong financial backing and meaningful long-term runway
Work alongside fast, product-minded engineers who ship at exceptional speed
Receive a highly competitive compensation package and equity
Location This is a full-time, in-person role in New York City. Candidates must already live in NYC or be willing to relocate.
#J-18808-Ljbffr
This is not a chatbot or experimental AI feature. The product combines conversational intelligence, live inventory and personalised recommendations to help consumers discover and book experiences through a single conversation.
You’ll join a small, exceptionally fast-moving engineering team and take end-to-end ownership of the product’s conversational intelligence layer. Your job is to make the system consistently produce accurate, useful actions—not simply convincing responses.
What You’ll Own
The complete agent architecture, including prompt systems, tool calling, multi-step reasoning and streaming
Conversational state, memory and preference-recall infrastructure
LLM-powered ranking, retrieval and personalised recommendations
Multi-turn reliability, including refinement loops, error recovery and graceful degradation
The rendering layer between agent outputs and the consumer-facing product
Model-provider strategy across platforms such as OpenAI, Anthropic and Gemini
Structured outputs, caching, routing and fallback behaviour
Evaluation infrastructure for agent, retrieval and recommendation quality
Investigation and resolution of real-world production failure modes
What We’re Looking For
3+ years of experience as a backend or full-stack software engineer
Experience building and shipping consumer-facing products—not solely internal tools or B2B software
Proven experience taking LLM-powered ranking, retrieval, search or recommendation systems into production
Experience building systems in which a model interprets user preferences and surfaces relevant results
Broad capability across the applied AI stack, including prompting, embeddings, retrieval, memory and personalisation
Practical experience defining, running and improving evaluations for ranking or retrieval quality
The ability to explain specific AI-system failure modes you have diagnosed in production
Strong product judgement and a focus on user outcomes
Regular use of AI-native tools and workflows in your own development process
Technology The primary application stack is TypeScript, although strong Python engineers are also encouraged to apply. Direct experience with every tool is not required; we care more about engineering fundamentals, learning speed and evidence that you can ship reliable AI products.
You’ll Thrive Here If You
Move quickly without lowering your quality bar
Enjoy owning ambiguous, technically difficult product problems
Think across the entire system rather than specialising in one narrow part of the AI stack
Test ideas by building and putting them in front of users
Care deeply about reliability, latency and the quality of the final experience
Have side projects or other evidence of independent, AI-native building
Why Join?
Own a critical product layer from architecture through production
Work on a genuinely agentic consumer experience with real-world utility
Join a small team with unusually high individual ownership
Build with strong financial backing and meaningful long-term runway
Work alongside fast, product-minded engineers who ship at exceptional speed
Receive a highly competitive compensation package and equity
Location This is a full-time, in-person role in New York City. Candidates must already live in NYC or be willing to relocate.
#J-18808-Ljbffr