Model Behavior, Product

The Browser Company

$200K — $275K *
US-AnywhereRemote in United States
Consumer Technology
Less than 5 years of experience
Job Overview by Ladders

Qualifications

  • 4+ years in product, design, or builder roles with experience in taking products from concept to release.
  • Ability to analyze ambiguous problems and design measurable evaluations to address them.
  • Hands-on experience with AI tools like Codex, Claude Code, or Cursor, and comfortable integrating new models.
  • Strong desire to contribute as an individual contributor, with a track record of frequent code submissions.
  • Proven impact in high-output teams, with a readiness to tackle complex challenges and learn in a fast-paced environment.

Responsibilities

  • Define and design the AI behavior for Dia based on user needs.
  • Continuously deploy changes weekly that directly affect user experiences.
  • Establish competency in evaluating and tracking quality improvements systematically.
  • Develop tools to enhance team workflows and faster iteration processes.
  • Stay updated and test new AI models and architectures as they become available.

Benefits

  • Comprehensive total compensation package including salary, equity, and benefits.
  • Wide range of perks supporting employee well-being, community engagement, and family needs.
  • Remote-friendly work environment with options to work from the New York office.
Full Job Description
About The Role

At the Browser Company, you will play a crucial role in shaping how Dia interacts with users. This is a hands-on role at the intersection of product and AI engineering: you will define Dia's intended behavior, ship prompt and code changes that deliver it, and own evals that measure quality and cost. The work spans the full stack of modern AI product development: prompting, tools, model selection, agent harnesses, and evaluation systems.

You'll report to our Model Behavior Lead and work closely with engineering and product. You'll have high impact and own problems end to end: understanding what users need, deciding the right approach, shipping the change, and proving it worked.

Overall you will...
  • Shape Dia's AI experience. Figure out what the right behavior is for users, then find the best path to it - a prompt change, a new tool, a different model or harness, better evals, or a spec that aligns the team.
  • Ship every week. This is a hands-on role: you'll be in the prompts, the code, and the data yourself, making changes that reach users continuously.
  • Improve systematically. Ground your work in evals that make quality trackable, so improvements hold up over time and the team can raise the bar faster with every iteration.
  • Build the tools that make us faster. Spot where the team's workflow is slow or manual and partner with engineering (or build it yourself) to create the internal tooling that lets us iterate at scale.
  • Keep us at the frontier. AI moves fast, and Dia has to move with it. You'll test new models as they ship, evaluate new harnesses, and prototype AI architecture strategies and push the ones that matter into how we build.

Recent examples of projects the Model Behavior team led:
  • Morning Brief. Prototyped our best performing feature of all time. The Morning Brief started as an idea from our Head of Engineering and was handed to Model Behavior to shape and bring to life. It required forming an opinion around what a morning brief should include, how to present it, and finally how to ensure quality was acceptable.
  • Chat architecture. Our move to an agent harness began as a prototype over our winter shutdown. It now powers Dia's chat, and it's the latest in a line of upgrades the team has led since GPT-3.5.
Qualifications
  • 4+ years of experience in product, design, or adjacent builder roles. You have hands-on experience taking a product from concept to its initial release (v1). Previous experience as a founder is a plus.
  • You think in systems. You can take an ambiguous problem, break it into something measurable, and design the evals to steer it. You can navigate when to decide on a short improvement fix versus investing in systematic fixes.
  • You work inside AI tools daily. Codex, Claude Code, or Cursor are core parts of your workflow, and you can articulate the reasoning behind your tooling choices. You try new models and harnesses as they ship, and fold what works into how you build.
  • You want to be an IC. This is a hands-on role where you'll be in the prompts, code, and data yourself. You will put up multiple PRs weekly.
  • You're here to make an impact. Model Behavior is a small, high-output team working at the core of all our AI features. We move fast and take on hard problems, which means you'll be learning constantly and growing quickly.
  • We're primarily focused on hiring in North American time zones and require that folks have 4+ hours of overlap time with team members in Eastern Time Zone.
Compensation and Benefits
  • Our total compensation package for full-time employees includes base salary, equity, and benefits. The annual salary range for this role is $200,000 - $275,000 USD. The actual salary offered will vary based on experience level and interview performance.
  • Benefits: We also offer a wide range of perks and benefits designed to support you, your family and to help you engage with your local community. To learn more, visit go.atlassian.com/perksandbenefits.
  • Location: We're a remote-friendly company and can hire in any country where Atlassian has a legal entity. If you live in New York (or want to visit), you're welcome to work from our beautiful office in Williamsburg.

Similar Jobs

More Jobs at The Browser Company

More Consumer Technology Jobs

Find similar Model Behavior, Product jobs: