Senior AI Product Engineer
- Role
- AI / ML
- Experience
- Senior
- Employment
- Full-time
Open to UTC-4…UTC+4. Set where you work from to check your eligibility.
No BS summary
Senior software engineer with 5+ years of experience, at least 2 building production LLM/AI agent systems. Must design agent architectures, build evaluation frameworks, and be comfortable across the stack. Remote within ±4 hours of GMT.
Optional skills
Job Title: Senior AI Product Engineer
Location: Remote +/- 4 hours GMT for collaboration
Type: Full-time, Permanent
Working style: Remote-first
About Kota
Kota is reimagining insurance and retirement benefits for the modern workforce. Through our two flagship products, Kota Platform, the first fully integrated benefits platform, and Kota Embed, our embedded insurance solution, we are making benefits more accessible, valued, and global.
Today, we power benefits for tens of thousands of employees at Europe's most forward-thinking companies, including Carwow, Tines, Zoe, and Remote. Founded in 2022, we have raised more than 20M EUR from leading investors like EQT Ventures, Eurazeo, Northzone, and Frontline, along with founders and executives from Workday, Remote, Personio, OpenAI, and more.
We work remotely and our team is spread across the world.
The Role
This is a role on Kota's AI pod. We are building out a generative AI capability from the ground up, the first of its kind at Kota, and we need a senior engineer who can help us do that with rigour and taste.
You will work directly with the AI Product Lead to design, build, and evaluate agent systems that help Kota's employer admins and employees make better use of their benefits data. This is early-stage, high-ownership work. You will be making real architectural decisions, not executing a predefined playbook.
What You'll Do
- Design and ship AI agent systems end to end, from early experiments through to production, owning the data access layer, retrieval strategy, and integration with Kota's existing infrastructure including Metabase and internal APIs
- Build and own the evaluation framework from scratch, including golden datasets, regression harnesses, and the ability to distinguish retrieval from generation failures; think adversarially about prompt injection, data boundary violations, and silent failure modes in a regulated context
- Set the technical direction for AI at Kota, with a seat at the table on agent architecture, and where AI creates the most leverage across the product
- Work across the pod and Platform team to surface new AI opportunities, collaborate with the AI Product Lead, and raise the bar on engineering practice through peer review and documentation
- Pick up slack and expand the role, this is an early pod with room to shape it; the remit will grow with you
What We're Looking For
Essential
- 5+ years of software engineering experience, with at least 2 years working directly on LLM or AI agent systems in production
- Strong grasp of agent architecture, including context management, tool use, retrieval strategies, and prompt design, and the ability to make and defend tradeoffs between them
- Experience building evaluation frameworks, not just using them: golden datasets, regression testing, and retrieval vs. generation failure analysis
- Comfort working across the stack
- Clear written communication; you can explain a technical decision to a non-technical stakeholder without losing precision
Strong plus
- Background in fintech, payroll, HR tech, or benefits; you understand why data freshness and auditability matter in regulated contexts
- Experience with AWS, Azure, or the broader Microsoft ecosystem
- Experience generating synthetic evaluation data to bootstrap eval coverage before real usage data is available
- Experience working in early-stage or scale-up environments where the architecture is still forming
What good looks like
You have built an AI system that failed in production and have a clear, unsentimental account of what went wrong, what you put in place to catch it earlier, and what you would do differently. You think about evals before you think about demos. You know the difference between a retrieval problem and a generation problem. You make deliberate decisions about what data an agent needs and when, rather than defaulting to retrieve-everything or hardcode-everything.
Our Values
- Have Integrity. We care deeply about our why and it will not jeopardise our how.
- Care Deeply. Our products are our story; we look at the details and go the extra mile to delight customers.
- Own Your Ship. Ask why, do the work, get the data, solve the problem, be an owner.
- Raise the Bar. We continuously look to push past good enough. Expect more, move faster, and never stop raising your and the team's standards.
- Play to Win. We exist because we decided to say no to the norm, and now we do this every day.
Kota Benefits
- Health Insurance via Kota
- Workplace Pension with matched contributions up to 5% via Kota (country dependent)
- WFH stipend to support your home office
- 25 days of Paid Time Off, work hard and recharge fully
- Annual company-wide offsite and dedicated team offsite
- Attractive stock options
We are an equal-opportunity employer and value diversity at our company. We do not discriminate based on race, religion, colour, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.
What you'll do
- Design and ship AI agent systems end to end, from early experiments through to production, owning the data access layer, retrieval strategy, and integration with Kota's existing infrastructure including Metabase and internal APIs
- Build and own the evaluation framework from scratch, including golden datasets, regression harnesses, and the ability to distinguish retrieval from generation failures; think adversarially about prompt injection, data boundary violations, and silent failure modes in a regulated context
- Set the technical direction for AI at Kota, with a seat at the table on agent architecture, and where AI creates the most leverage across the product
- Work across the pod and Platform team to surface new AI opportunities, collaborate with the AI Product Lead, and raise the bar on engineering practice through peer review and documentation
- Pick up slack and expand the role, this is an early pod with room to shape it; the remit will grow with you
What they require
- 5+ years of software engineering experience with LLM and AI agent systems in production
- Strong grasp and depth in agent architecture, including context management, tool use, retrieval strategies, and prompt design, and ability to challenge and defend tradeoffs across these
- Experience building evaluation frameworks, not just using them: golden datasets, regression testing, and retrieval vs. generation failure analysis
- Comfort working across the stack
- Clear written communication, able to explain a technical decision to a non-technical stakeholder without losing precision
- Preferred: Background in fintech, payroll, HR tech, or benefits and understanding why data freshness and auditability matter in regulated contexts
- Preferred: Experience with the AWS, Azure, or broader Microsoft ecosystem
- Preferred: Experience generating synthetic evaluation data to bootstrap eval coverage before real usage data is available
- Preferred: Experience working in early-stage or scale-up environments where the architecture is still forming
Benefits
- Health Insurance via Kota
- Workplace Pension with matched contributions up to 5% via Kota (country dependent)
- WFH stipend to support your home office
- 25 days of Paid Time Off, work hard and recharge fully
- Annual company-wide offsite and dedicated team offsite
- Attractive stock options
Kota is making insurance and retirement benefits as accessible as modern software through the Kota Platform and Kota Embed, an embedded insurance and retirement product for platforms.