About Chaseit
Chaseit builds AI voice agents for loan servicing and collections. Our agents make tens of thousands of human-like calls every day for lenders across Europe and beyond. They handle everything from payment reminders to payment-plan negotiation, in 10+ languages, while staying compliant and genuinely empathetic.
We're a VC-backed startup with live customers, real revenue, and a lean team where every person has outsized impact.
What you will do
Every live agent we run gets feedback. A client flags a call where the agent sounded too pushy with someone who'd just lost their job. Or the agent misunderstood a question about a payment date. Or it skipped a line the lender's compliance team needs said. You'll own what happens next: working out what went wrong and changing the agent so it doesn't happen again.
You'll work closely with our Deployment Strategists, who own the client relationships. You'll be the person who turns what clients tell us into better conversations.
You will:
- Own incoming client feedback on live voice agents: review the flagged calls and transcripts, work out what actually went wrong, and decide on the fix
- Update prompts and conversation flows to address it by rewording, restructuring, adding rules, or handling new edge cases, without breaking what already works
- Create prompts for new agents, use cases, and clients, based on their scripts, policies, compliance requirements, and tone of voice
- Test your changes before they go live: call the agent yourself, walk through the tricky scenarios, and check that fixing one thing didn't break another
- Keep a clear record of what changed and why, and close the loop with the Deployment Strategist or client so they know it's been handled
- Notice when the same issue keeps showing up across clients and flag it to the team
If you want to go further, there's plenty of room. You could propose ideas that make our agents sound more natural, build test scenarios and QA checklists, hunt for edge cases before clients find them, or help us evaluate changes more systematically. None of that is required on day one, but we'd love it.
Why this role exists
Fixing issues as clients raise them is real, important work. It keeps quality high and it's how clients learn to trust us. As we deploy more agents for more lenders in more languages, the volume of feedback keeps growing. Doing it well now takes someone who cares about it full-time.
It's also more of a craft than it looks. Feedback like "the agent sounds robotic" or "it gave up too easily" rarely traces back to one line in a prompt. You'll need to listen carefully, find the real cause, and make a precise change that fixes the problem without creating two new ones.
Who you are
- Hands-on experience writing prompts for LLMs, ideally in a real product. You've watched a prompt fail and figured out how to fix it
- Good experience in Python: writing prototyping & testing scripts, analysing call data.
- Excellent written English. Prompts are writing, and clear, precise writing makes better agents
- A good ear for conversation: you notice when something sounds off (robotic, pushy, confusing, too long) and can explain why
- Careful and methodical: you test your changes instead of assuming they work, and you catch the details others miss
- Comfortable working with clients and colleagues: you can turn vague feedback into a concrete problem and ask the right questions when it's unclear
- Ownership: you take a piece of feedback from "reported" to "fixed and confirmed" without anyone chasing you
- Comfortable in a fast-moving startup where priorities shift and some days are firefighting days
Nice to have
- Background in conversation design for chatbots, IVR, or voice assistants
- QA or testing experience, especially with conversational AI
- Experience in a call center, customer service, or collections, so you know what a good call sounds like
- Additional European languages (our agents speak 10+)
- Familiarity with lending, collections, or another regulated industry
What you'll get
- Salary: €1500 – €4000 gross per month, depending on experience
- Direct work with the founding team, and a real say in how our agents talk to thousands of people every day
- Room to grow: as you take on more (QA, evaluations, experiments), the role grows with you
- A flexible schedule between the office and home
- Lunch covered whenever you're in the office
Location: Vilnius, Lithuania. We work on-site with a flexible office/home schedule, so you'll need to be based in Vilnius or able to relocate there.