Full Stack AI Engineer | Onsite in New York | Local Candidates Only, 5

Our client is looking for a highly hands-on engineer who can help build and launch AI-native products from the ground up. You will work closely with the founders to turn ideas into production-ready applications and help shape the company’s technical foundation., * Build, test, launch, and improve full-stack products from zero to one

• Develop modern frontend applications using React, Next.js, or similar technologies
• Build backend services, APIs, databases, and third-party integrations
• Create AI-powered features using large language models and multimodal models
• Build AI agents that can use tools, complete workflows, and interact with external systems
• Develop retrieval-augmented generation, embeddings, vector search, and knowledge systems
• Integrate voice AI, speech-to-text, text-to-speech, or real-time conversational systems
• Implement structured outputs, tool calling, model routing, and human approval workflows
• Build evaluation, monitoring, and testing systems for AI performance
• Improve application reliability, scalability, security, latency, and AI inference costs
• Work directly with founders, users, and product stakeholders
• Help make product, architecture, and technology decisions

The right candidate is comfortable working across frontend, backend, infrastructure, and AI systems. You should enjoy fast-moving startup environments, changing requirements, rapid iteration, and taking complete ownership of what you build., * 4 or more years of professional software engineering experience

• Proven experience building and launching a product from zero to one
• Prior experience working at an early-stage startup
• Strong experience across frontend and backend development
• Strong skills in TypeScript, JavaScript, React, and Next.js
• Backend experience with Node.js, Python, FastAPI, Go, or similar technologies
• Experience with PostgreSQL, APIs, cloud infrastructure, and production deployments
• Hands-on experience integrating large language models into production applications
• Experience with AI agents, RAG, tool calling, embeddings, multimodal AI, or voice AI
• Strong product judgment and the ability to operate with limited direction
• Ability to move quickly while maintaining high engineering standards
• Must be actively hands-on and writing production code

Preferred Experience

• OpenAI, Anthropic, Gemini, or open-source AI models
• Model Context Protocol and agent orchestration frameworks
• LangGraph, LlamaIndex, Vercel AI SDK, or similar tools
• Vector databases such as Pinecone, Qdrant, Weaviate, or pgvector
• Voice AI, WebSockets, WebRTC, or real-time streaming systems
• AI evaluations, observability, prompt injection protection, and model security
• Docker, AWS, Google Cloud, Azure, or serverless infrastructure
• Experience as an early, founding, or first engineering hire

Ideal Candidate

You are a builder who prefers creating new products over maintaining mature systems. You are comfortable wearing multiple hats, making decisions quickly, and turning incomplete ideas into products customers can use.

You understand that strong AI products require more than connecting an application to a model API. You know how to build dependable workflows, evaluate output quality, manage latency and cost, and create safeguards around AI-generated actions.

Candidates whose experience is exclusively with FAANG companies, large enterprises, or mature technology organizations will not be considered. FAANG experience is welcome when combined with meaningful early-stage startup and zero-to-one product experience.

Location Requirement

This position is onsite in Manhattan five days per week.

Candidates must already live within a reasonable commuting distance of New York City. Remote, hybrid, relocation-dependent, and out-of-area candidates will not be considered.