Agent Engineer

About Conduit

Conduit is an AI agent platform purpose-built for hospitality. Hotels, vacation rentals, and resorts use our agents to automate millions of workflows across guest communication and operations, all to give their guests a better experience.

Hospitality is 10% of global GDP and it turns on customer experience. Resolving a guest request immediately is most of that experience, and resolving one means moving across several tools and between several people. That coordination is the work, and it is what our agents do.

Role: Agent Engineer

You build, configure, and evaluate the agents that run guest operations for our customers. We are hiring several agent engineers. Each one owns the agents for a set of customers, hotels and property managers, end to end: guest communication, operations, review management, and reporting.

This is a forward-deployed role, and mostly an operational one. Most of your time goes into building agent workflows that scale and evaluating them. You will often join calls with operators to resolve issues, but most of the job is the agents themselves.

You report to our Head of Operations and work closely with Punn, our CTO and co-founder, and the engineering team.

Why this role is different

• You work from real conversations. Guest transcripts are messy, and finding the failure in them is the job.

• You change production agents from your first day.

• You test changes with evals, and you use your judgment to know when a fix needs one.

What the work looks like

• Reading transcripts and escalations to find where an agent got it wrong, then fixing the prompt, the tool, or the workflow.

• Building scalable workflows across communication, operations, review management, and reporting.

• Building and maintaining eval sets for each agent and each customer.

• Joining calls with operators to resolve issues, and turning how they work into agent behavior.

• Working with our engineers when an agent needs a new capability in the platform.

Who you are

• You write production code in Python or TypeScript.

• You have configured AI agents that run in production.

• You have run evals on those agents.

• You can explain a fix in plain words, and you can also go deep into the details.

• No hospitality experience needed.

• This role is onsite in San Francisco or remote. Candidates must be authorized to work in the US; we generally do not sponsor visas.

Compensation and Benefits

• $150K to $230K base

• 0.05% to 0.25% equity

• Comprehensive health insurance and 401k match

• Dinner covered every day when you work from our San Francisco office

Interview Process

• Intro call with Punn.

• Technical screen with Punn. Code, data handling, and tool design for agents.

• Agent round with Yash.

• Final round with Punn or Cole.

• References, then an offer.