Applied AI Engineer, Product Agents

Location: 📍 In-office, 5 days a week at our SoHo office in New York City. We believe great things happen when we build side by side.

Why This Role

Henry is the AI analyst for commercial real estate. Eight of the ten largest enterprise CRE firms already run on Henry, and we’re scaling those deployments across their brokerages, research teams, and capital markets groups. We recently raised our Series A at a $120M valuation. Demand is not the problem. Building fast enough to meet it is. The product is an analyst in a box. Henry reads the email, builds the Excel, writes the PowerPoint, and keeps the property database current, all on one platform. Work that used to take a junior analyst a week now happens in minutes, inside the tools CRE teams already live in. We’re a small team shipping agents into real enterprise workflows. We’re looking for engineers who want to define what that looks like as the models and paradigms keep changing underneath us.

Responsibilities

• Build agentic products: Design and ship multi-agent systems that orchestrate across email, spreadsheets, and decks, manage large and messy context at scale, and get real work done for analysts.
• Evaluate and hill climb: Build both subjective and programmatic evals, decide what they should measure, and use them to hill climb on agent quality until customers are happy. “It works on my laptop” is where the job starts, not where it ends.
• Own the product end to end: Work directly with stakeholders across the company. Triage customer feedback from Customer Success, sit with users to understand what they actually need, and take the fix from design through build and deployment without waiting for a spec.
• Automate how we learn from users: Build systems that automatically detect subjective user feedback (frustration, rework, abandonment, silent workarounds) so we hear about problems before customers tell us.
• Experiment constantly but judge honestly: We’re highly experimental and willing to try anything, including new models, harnesses, and agent frameworks the day they come out. We’re also grounded in reality. You’ll measure whether something actually moved quality, and kill it if it didn’t.

How We Assess

• You’ve shipped something with an agent at its core: You’ve built a product or automated a real workflow where an LLM agent does the work. You have hands-on experience with multi-agent orchestration, context management at scale, and both programmatic and subjective evals.
• You try new AI tools by default: You’ve used the latest model or agent framework before anyone asks, and you have an opinion on whether it’s worth anything. You re-test old conclusions when a new release comes out, because “that didn’t work” has an expiration date.
• You can defend your design choices: You can explain why you picked your harness, how your eval framework tells you the system is working, and how you manage context: what goes in, what stays out, and why.
• You own problems end to end: You start by sitting with the user to find what they actually need, then take it through design, build, and deployment without handing off halfway.
• You turn deep expertise into agent requirements: You can sit down with someone who has spent 30 years underwriting an asset class you’ve never touched, pull out how they think, and turn that into clear requirements and evals for an agent. You don’t need to know CRE today, but you need to learn it fast from the people who do.
• You’re experimental and grounded at the same time: You’ll try anything once, and you’ll be the first to say it didn’t work.

What We Offer

• 💰 Competitive salary + equity: A top-of-market base and meaningful stock options, so you share in what we build
• 🩺 Health, dental & vision: Comprehensive coverage for you and your family
• 🏋️‍♂️ Health & wellness stipend: Stay sharp with access to Equinox gyms, on us
• 🌴 Generous PTO: Take the time you need to recharge, plus company holidays
• 🥗 Breakfast, lunch & dinner: Catered meals in the office, every day
• 🏙️ Beautiful SoHo office: Work from a bright, inspiring space in the heart of Manhattan
• 🚀 Early-stage impact: Work directly with the founders, own real problems, and help shape the company from the ground up

Notice to Candidates

To ensure fairness and proper consideration, we do not accept resumes or expressions of interest via email or social media messages. If you’re interested in a role, please submit your application directly through our careers page.

Please note that Henry AI may use AI notetakers and other recording devices in the recruiting process. If you interview with us, with your consent, we may record your conversations and summarize them into notes for internal use. Recording is optional, and declining will not affect your candidacy.

Henry AI is an equal opportunity employer. We are committed to diversity and inclusion in our company. We do not discriminate based on race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.