Application Guide

How to Apply for Tech Lead, AI Safety

at Trajectory Labs, PBC

🏢 About Trajectory Labs, PBC

Trajectory Labs, PBC is an early-stage startup building RL environments for frontier AI labs to train robust, secure, and reliable models. Unlike big labs, you'll work directly with cutting-edge partners and shape safety methodology from the ground up in a high-autonomy, low-structure environment.

About This Role

As Tech Lead for AI Safety, you'll own the design and execution of safety and alignment evaluations, red-teaming programs, and RL environments that directly diagnose and improve frontier model behavior. You'll lead a small team, partner with frontier AI labs, and use coding agents and LLMs as core development tools—making autonomous technical and operational decisions in a fast-shifting startup context.

💡 A Day in the Life

You might start by reviewing a coding agent's overnight run that generated adversarial prompts for a frontier model, then meet with your team to refine the evaluation methodology. Later, you'll sync with a partner lab to align on red-teaming goals, and spend the afternoon prototyping a new RL environment using LLM-assisted coding. Expect frequent context switches and autonomous prioritization.

🚀 Application Tools

Generate Cover Letter

AI-powered, tailored for Trajectory Labs, PBC

Interview Prep

Practice questions for this role

🎯 Who Trajectory Labs, PBC Is Looking For

  • Extensive hands-on experience with coding agents (e.g., SWE-agent, OpenHands, Aider, or custom agent frameworks) and LLMs for development, not just casual use.
  • Proven track record designing safety or alignment evaluations, red-teaming programs, or RL environments for large language models.
  • Comfortable leading a small team in an unstructured, early-stage startup where priorities shift and you must own methodology, architecture, and results end-to-end.
  • Strong background in AI safety/alignment (e.g., adversarial robustness, deception detection, scalable oversight) and ability to translate research into production-grade evaluations.

📝 Tips for Applying to Trajectory Labs, PBC

1

Showcase concrete projects where you used coding agents or LLMs as core development tools—link to repos, demos, or detailed write-ups that demonstrate autonomous decision-making.

2

Highlight any experience designing safety evaluations or red-teaming programs for frontier models; quantify impact (e.g., number of vulnerabilities found, adoption by labs).

3

Emphasize your ability to operate with minimal structure: describe a time you set your own roadmap, made architectural decisions, and shipped under shifting priorities.

4

Mention specific frontier AI labs you've partnered with or would like to partner with, and how you'd approach collaboration on safety evaluations.

5

Avoid generic cover letters; instead, propose a concrete evaluation or RL environment idea that Trajectory Labs could build for a frontier lab, showing you understand their business.

✉️ What to Emphasize in Your Cover Letter

['Your hands-on experience building with coding agents and LLMs, with specific examples of how you used them to accelerate safety research or evaluation development.', 'A concrete proposal for a safety evaluation or red-teaming program you could lead at Trajectory Labs, including methodology and expected outcomes.', 'Evidence of leading technical teams in ambiguous, early-stage settings—how you balance autonomy, speed, and rigor.', 'Your motivation for joining a public benefit corporation focused on AI safety, and why partnering with frontier labs excites you.']

Generate Cover Letter →

🔍 Research Before Applying

To stand out, make sure you've researched:

  • → Explore Trajectory Labs' website and any public materials to understand their specific approach to building RL environments for frontier AI labs.
  • → Research recent papers or blog posts on AI safety evaluations and red-teaming from organizations like Anthropic, OpenAI, DeepMind, and ARC Evals to speak the same language.
  • → Investigate the current landscape of coding agents (e.g., SWE-agent, Devin, OpenHands) and how they are used in AI safety research to demonstrate up-to-date knowledge.
  • → Look into the public benefit corporation (PBC) structure and what it means for a startup's mission and culture, so you can articulate why it matters to you.
Visit Trajectory Labs, PBC's Website →

💬 Prepare for These Interview Topics

Based on this role, you may be asked about:

1 How would you design an RL environment to train a frontier model to be more robust against adversarial prompts? Walk us through your approach.
2 Describe a time you used coding agents to automate part of an evaluation pipeline. What were the trade-offs and results?
3 How do you prioritize safety evaluations when working with a frontier lab that has competing deadlines and shifting priorities?
4 What metrics or signals do you use to diagnose model behavior in red-teaming exercises, and how do you ensure they generalize?
5 How would you lead a small team of engineers and researchers to deliver rigorous safety evaluations in a low-structure startup environment?
Practice Interview Questions →

⚠️ Common Mistakes to Avoid

  • Focusing only on high-level AI safety philosophy without demonstrating hands-on experience building evaluations or RL environments.
  • Underestimating the 'coding agents' requirement—treating it as a buzzword rather than a core tool you've deeply integrated into your workflow.
  • Showing discomfort with ambiguity or needing clear structure; this role requires proactive decision-making in a shifting startup environment.

📅 Application Timeline

This position is open until filled. However, we recommend applying as soon as possible as roles at mission-driven organizations tend to fill quickly.

Typical hiring timeline:

1

Application Review

1-2 weeks

2

Initial Screening

Phone call or written assessment

3

Interviews

1-2 rounds, usually virtual

✓

Offer

Congratulations!

Ready to Apply?

Good luck with your application to Trajectory Labs, PBC!

← Back to Job Listing Apply Now →