About the Role
Fini is building the future of enterprise customer support. Our self-learning AI agents autonomously resolve up to 90% of customer tickets, lift CSAT by double digits, and cut support costs at scale, for companies like Bitdefender, TrainingPeaks, and CoverGenius.
Our agents improve on their own. But getting a new enterprise to production-grade reliability is an active discipline: defining what "correct" looks like for their world, then using our self-learning tooling to turn that signal into a better agent, faster than anyone else can. That's the work.
We're looking for an AI Solutions Engineer who is rigorous, curious, and obsessed with getting AI behavior right. The role sits at the intersection of AI reasoning, customer solutions, and real-world performance. You'll work directly with our customer and engineering team to operate Fini's self-learning loop, build the evaluation signal that powers it, and help define what "good" looks like at enterprise scale.
What You Will Own
Self-Learning Operations
- Use Fini's self-learning tooling to run and tune the improvement loop for enterprise agents.
- Turn evaluation signal into measurable agent improvements, and verify each iteration moves the right metrics.
- Build repeatable workflows so the Solutions team can drive this loop at scale.
Evaluation & Golden Sets
- Design and run structured evaluations across Fini's AI agents for enterprise customers.
- Maintain and expand golden sets, the curated conversation benchmarks that define ground truth and feed the learning loop.
- Track evaluation metrics over time and flag regressions as customer configurations evolve.
AI Reasoning Analysis
- Trace why the model responds the way it does: reasoning paths, prompt sensitivity, knowledge boundaries.
- Surface the edge cases that teach the agent to improve, and categorize behavior across ticket types, industries, and configurations.
- Turn those findings into insights that directly inform agent improvements and customer outcomes.
What We're Looking For
- Strong analytical thinking. You enjoy pulling problems apart and finding root causes.
- Comfort with ambiguity. You can form a hypothesis when there's no clear answer.
- Attention to detail that borders on obsessive.
- Strong familiarity with LLMs and prompt engineering. You treat AI as a co-worker.
- Experience with structured data analysis and Python for parsing, API calls, or prompt templating, with or without AI helping you write it.
Bonus Points
- Exposure to customer support, CX, or enterprise SaaS.
What Success Looks Like
- You drive the self-learning loop to measurably better agents, iteration over iteration.
- Golden sets that catch regressions before they reach customers.
- Attention to detail that borders on obsessive.
- Clear, documented behavior taxonomies that make the Solutions team faster.
- Enterprise customers going live with AI agents that are accurate, reliable, and trusted.
- You understand Fini's AI better than almost anyone, and can explain exactly why it behaves the way it does.
Why Join Fini
- You'll work on AI that's in production, handling millions of real customer conversations.
- You'll operate self-learning infrastructure firsthand, a rare skill almost no one has yet.
- You'll work directly with a founding team backed by YC and Matrix, moving fast on hard problems.
Logistics
How to Apply
Send your LinkedIn, resume, and a short note on what interests you about AI and customer solutions to careers@usefini.com, subject line "AI Solutions Engineerat Fini." We move fast, so will you hear back quickly.
