What are the responsibilities and job description for the Software Engineer position at Prove AI?
About Prove AI
We're building tools that make multi agent systems reliable, de-buggable, and fast to build for engineers and nontechnical people alike.
We just rebooted the engineering team and are starting fresh with a new direction. Small team, clean codebase, real funding, and a product thesis we're excited about.
About the Role
You'll be one of the first engineers. You'll shape the product, the architecture, and the culture. There's no ticket queue. You'll help figure out what to build and why, then build it.
You'll spend as much time reasoning about what to build and how systems should fit together as you will writing code. A lot of the code gets written by AI. Your job is to know what good looks like, catch what the model misses, and steer it toward the right outcome.
What we care about most
Verbal reasoning.
You can take a messy technical problem, think through it out loud, and land on a framing that moves the conversation forward. You explain trade-off's clearly to engineers, nontechnical teammates, and AI agents. Meetings with you are productive, not performative.
Systems thinking.
You understand how software systems actually work end to end: how services talk to each other, where failures hide, why something that looks healthy can be making bad decisions underneath. You don't need deep expertise in every layer, but you need a mental model of the whole picture. That's what lets you guide AI coding tools well and catch when they're building the wrong thing confidently.
Curiosity about how AI is changing software.
Not just agents specifically, but the whole shift: how development works when AI writes most of the code, what breaks when systems make decisions at runtime instead of design time, why the tooling and processes we inherited don't quite fit anymore. You have opinions about where this is going.
Speed and judgment.
You ship fast but you're not reckless. You know when to prototype and when to build for real. You don't over engineer and you don't under think.
Ownership.
This is a start-up. Things will be ambiguous. You're energized by that, not paralyzed by it.
Minimum requirements
- Working proficiency in Python or TypeScript.
- Enough systems knowledge to reason about distributed behaviour from logs, traces, and metrics.
- Familiarity with Docker and containerized services.
- Some exposure to LLMs, whether through building with them, prompting them seriously, or studying how they work.
- Ability to use AI coding tools effectively, meaning you know when to trust them, when to override them, and how to specify what you actually want.
Helpful but not required
- Experience with observability or monitoring infrastructure.
- Background in developer tools or platform engineering.
- Exposure to agent frameworks or LLM application development.
What we don't care about
- Years of experience.
- Specific languages or frameworks.
- Where you went to school.
- Whether your resume has the right keywords.
What You’ll Do
The core problem: when AI agents make decisions in production, nobody can tell you whether those decisions were good, which one caused a bad outcome, or what to change. We're building the tooling layer that answers those questions. That means observability infrastructure, evaluation systems, attribution, policy enforcement, and interfaces that make complex agent behavior legible. What you work on will change as we learn. The problem is stable. The solutions aren't yet.
- Competitive salary plus significant equity.
- Well capitalized, this isn't a 12 month runway situation.
- Work directly with senior product and engineering leadership
- Build production systems that AI teams depend on daily
- ESOP equity
- Flexible hours
- Generous PTO
- Global offsites
- Education support
- Clear advancement path