About this role.
RevenueCat is hiring a senior software engineer to build foundational infrastructure for an AI agent that helps developers analyze, debug, and grow subscription revenue. The role focuses on LLM orchestration, tool execution, structured outputs, context management, evaluation pipelines, observability, and trust or permissions controls. The engineer will ship agent features, own a core infrastructure area, improve reliability and performance, participate in on-call response, and contribute to system architecture. This is a highly autonomous remote role for someone with strong backend engineering fundamentals and hands-on production LLM systems experience.
Role DNA
A quick view of the complexity, pace, ownership and collaboration implied by the job description.
Job Complexity
5/5Pace & Pressure
5/5Autonomy Level
5/5Communication Load
4/5Salary analysis
Estimated compensation compared with the broader US market for similar roles.
Core skills
Skills and capabilities most closely associated with this opportunity.
Sample interview questions
I would explain the user workflow, model and tool boundaries, orchestration approach, state or context strategy, and the controls used for failures. I would also quantify outcomes such as task success rate, latency, adoption, or reduction in manual support work.
I would use explicit tool schemas, least-privilege scoped credentials, authorization checks at the tool layer, input validation, audit logs, and clear separation between read and write operations. Higher-risk actions would require confirmation, policy evaluation, and potentially human approval before execution.
I would combine offline test-set measures such as task completion, tool-selection accuracy, groundedness, and regression rate with online measures such as successful resolution rate, latency, retry rate, user feedback, and escalation frequency. I would segment results by workflow and monitor failures through traces and structured logs.
I begin by clarifying the user problem, constraints, and success metrics, then map the existing system and identify the smallest high-confidence milestone. I document assumptions, propose alternatives with tradeoffs, align stakeholders early, and iteratively evolve the design based on production evidence.
I would describe how I used telemetry to isolate the bottleneck, formed and tested hypotheses, implemented a targeted fix, and added monitoring to prevent recurrence. A strong example would include measurable improvements in latency, error rate, throughput, or incident frequency.
RevenueCat removes the headaches of building and scaling in‑app subscriptions. Since graduating from YC’s S18 batch we’ve grown into the default monetization platform for mobile: we’re in >40% of newly shipped subscription apps, we process $12B+ in annual purchase volume, and we help everyone from a solo dev in Brazil to the OpenAI mobile team understand and grow their revenue.
We’re a remote‑first crew of 150+, spread across 25+ countries, and guided by values we actually practice: Customer Obsession, Always Be Shipping, Own It, and Balance. If you want your work to touch hundreds of millions of end‑users (and help the developers behind them get paid), you’ll fit right in.
The Role
We’re building an AI assistant that helps developers understand and grow their subscription revenue. It lives in Slack today and is expanding to our dashboard and API. It can pull real-time data, evaluate experiments, debug integrations, and explain what’s happening in your app using natural language. We see a future where the majority of a user’s interactions are either intermediated or automated by our agent.
We’re a team that’s historically built deterministic CRUD systems, and we need an engineer who can build the foundational agent infrastructure that everything else runs on: orchestration, tool execution, context management, evaluation, and the trust layer that lets us give an agent increasing autonomy over things that matter.
About you:
You have 5+ years of experience shipping production systems.
You have hands-on experience building with LLMs — not just prompting, but building the systems around them: tool use, structured output, context management, evaluation, orchestration.
You have strong backend fundamentals. You’ve built systems that need to be reliable, observable, and secure.
You are comfortable with ambiguity. This is an early-stage product inside a growth-stage company. The architecture is still being figured out.
You are self-directed. You figure out what needs to be built, build it, and ship it.
You collaborate well with others and can communicate effectively in a fully-remote culture.
Bonus:
Experience with OpenAI/Anthropic APIs specifically.
You’ve built agents that other people use.
Experience with evaluation and observability frameworks (Langfuse, Langchain, etc).
Familiarity with subscription business models, app stores, or developer tools.
You’ve built Slack integrations or other conversational interfaces.
Contributions to open-source AI/ML tooling.
In the first month, you’ll:
Ship your first agent feature and get familiar with the architecture, tool ecosystem, and how Rico talks to RevenueCat’s APIs.
Identify the biggest reliability or performance gap and start fixing it.
Meet with the team, get set up with repos, dev environment, and debugging tools.
Familiarize yourself with RevenueCat dashboards, logging, debugging tools, cloud providers, infrastructure management and general architecture.
Within the first 3 months, you’ll:
Launch a brand new agentic feature to tens of thousands of developers.
Own a core infrastructure area — orchestration, eval pipeline, tool framework, or the trust/permissions layer.
Be able to scope and work on projects self-sufficiently.
Learn the basics of incident response and be part of the on-call rotation.
Within the first 6 months, you’ll:
The agent is measurably more reliable, faster, and capable because you’re here.
You’ve built infrastructure that the rest of the team ships agents with.
Review code, create proposals, and contribute to architectural discussions.
Have shipped a major product or feature.
Within the first 12 months, you’ll:
RevenueCat is an AI-native company. You built the engine underneath it.
Know all the major components of our agent system and be able to debug complex issues.
Have your own initiatives for improving our agent products, understanding the current issues and priorities.
Mentor other engineers joining the team.
What we offer:
Competitive equity in a fast-growing, Series C startup backed by top-tier investors, including Y Combinator
10-year window to exercise vested equity options
Fully remote and flexible work environment
4-5 weeks of suggested time off annually for mental, physical, and emotional recharge
$2,000 USD for workspace setup and $1,000 USD annual stipend for continuous learning
Curious about the interview process? Discover more in our blog post about how we hire and learn tips to help you succeed.
Upload your resume now
- To unlock remote work opportunities and be discovered by global employers.
This job listing has been manually reviewed by the Jobicy Trust & Safety Team for compliance with our posting guidelines, including verification of the company's legitimacy, accuracy of job details, clarity of remote work policy, and absence of misleading or fraudulent content.
