AI Product Engineer

Neho is one of Switzerland's leading real estate brokerage companies, redefining the industry through innovation. We combine human expertise with powerful digital solutions to deliver a faster, more transparent, and more efficient real estate experience. Our approach has been recognised with prestigious awards, including the SEF Award 2023. What truly sets us apart is our fixed-price model, offering clients a modern alternative to traditional commission-based services, powered by advanced technology.

We are looking for an AI product engineer to lead, hands-on, how we put AI to work. You will build the agents that take the manual, repetitive parts of buying, financing and selling a home off people's plates, so our experts can spend their time on the moments that matter.

You will not start from zero. We have a compelling vision and working prototypes. Your job is to take this pattern from prototype to product: real agents, real users, measured outcomes. You will get a first agent into production within your first months and grow a pipeline from there.

This is a zero-to-one role with real ownership and visibility reporting directly to our CTO. If you love turning a manual process into working software that ships in weeks rather than quarters, you will feel at home here.

Workload: 80 - 100%

• Find the value: Spend real time with product managers, advisors, real estate agents and operations specialists, map their workflows, identify the agent use cases that remove the most measurable toil across the client journey or internal operations - prioritise by outcome

• Build the agents, hands-on: Design, build, test and ship agents yourself: use agent frameworks & platforms where they are faster, real code where they are not

• Own the delivery surfaces: Take our prototypes to product, and evolve how agents reach users: in-product, in chat, and later on mobile 

• Build the MCP layer: Wrap the stable domain APIs in thin, well-designed MCP tools your agents act through. You are the first and most demanding consumer of our API-first architecture.

• Drive the requirements to the teams: Specify the APIs your agents and surfaces need from the domain-owning engineering teams, who build and expose them; you consume clean contracts rather than reaching into systems directly

• Choose the platform with Platform: Lead the selection of the technology & platforms jointly with the Platform team, accounting for security, Swiss/EU data protection, cost and operability

• Ship on rails, within the guardrails: Deploy on the company's fenced, containerised, API-only path; instrument and evaluate your agents; respect data-sensitivity tiers

• Hand off when an agent grows up: When an agent becomes revenue-critical or touches critical client or financial data, hand it to Engineering for hardening and long-term ownership, and keep building the next one

• Measure outcomes: Track effort removed, cycle-time and conversion impact, and adoption by the people the agents serve. Not the number of agents shipped
  

• Not the builder of core APIs: Domain engineering teams own and expose the stable APIs; you drive the requirements, build the thin MCP wrappers on top, and consume the contracts

• Not the platform/infra owner: Platform owns runtime, deployment, observability and security; you build on top

• Not an AI services desk: You are not here to build bespoke tools for every department; you build the agents that move our priorities, and help others self-serve safely on the sanctioned path

• End-to-end ownership of your product with the authority to drive implementation, challenge the status quo, and own the outcome from conception to launch of new features

• A unique career opportunity to join a fast-growing and award-winning scale-up within the real estate industry

• Shape the future of real estate tech in Switzerland: Have a direct impact on Neho's growth, and deliver efficient, and scalable AI tools that give Neho a competitive edge

• Enjoy flat hierarchies, quick decision-making processes, and the freedom to work autonomously and leave your mark

• Be part of an energetic, forward-thinking team spanning our offices in Lausanne and Zurich as well as our remote engineering team in Romania

• Thrive in an entrepreneurial, innovative culture that values transparency, fairness, and strong team support

• Competitive benefits including 5 weeks of vacation + extra days based on tenure, fully paid maternity/paternity leave, remote work options, regular team events, and more!

• Fluent in the AI stack: You've worked with the modern AI stack in some form (direct model APIs, agent frameworks like LangGraph/CrewAI/Google ADK, orchestration platforms like Dify/n8n). We care about what you shipped, not which tool you used.

• A developer at heart: You have a software engineering background and have been writing and owning production code for years. Which specific language doesn't matter to us.

• Comfortable across the stack: Whether that's PHP, TypeScript/JavaScript, or Python, you're comfortable building applications, APIs and random other things like browser extensions or even mobile apps. 

• Pragmatic about tools: You reach for agent platforms and AI tooling where they're faster, and write real code where they're not. Comfortable with MCP, webhooks, prompting and streaming UIs.

• Disciplined about evaluation: You have shipped LLM-based systems to real users: you write evals, design quadrails, think in failure modes, and know when an agent must refuse to answer. In a regulated, document-heavy domain this is the core skill, not a nice extra

• A discovery person: You are comfortable sitting next to an advisor for an afternoon, mapping a messy human workflow, and earning the trust of the people whose toil you are removing

• Product-minded: You prioritise by outcome, validate before building, and are willing to kill an agent that does not move a metric

• Fast and pragmatic: You can get a first useful agent into production in weeks and iterate.

• Safety-aware: You respect data sensitivity (client, financial, identity) and work within guardrails rather than around them

• Language: Strong communication skills in English, French or German is considered a plus

• Experience with real estate, mortgages, fintech or another regulated, document-heavy domain

• Familiarity with Swiss/EU data-protection expectations

• Mobile development experience (for the later delivery surfaces)

• Month 3: first agent live in production, with measured effort removed.

• Month 6: a handful of agents in production across the journey and internal ops, delivered through a central delivery system and used daily by agents, ops or advisors; the platform choice settled with the Platform team; a clear, prioritised backlog.

• Month 12: agents are a normal, trusted way we remove toil; the hand-off path to Engineering is working; the case is made for a small agent pod if volume warrants it.

If this sounds like the right professional opportunity for you, then apply now with your CV in English.

Please note that we are only considering applications submitted through our career page: https://careers.neho.ch/