Blackletter is reinventing legal services from first principles. We pair frontier automation with top-tier lawyers to deliver elite legal services with fast turnaround, predictable pricing, and a better experience for clients and the lawyers doing the work. We’re serving a growing roster of clients across venture financing, commercial, corporate, and employment, and we’re backed by leading venture investors. Our founding team brings experience from frontier AI labs, leading law firms, and top law schools in North America.
We’re a small team making our earliest hires. You’ll have the opportunity to shape not only our company, but how legal work gets done. The Role We’re looking for a Machine Learning Engineer to own and improve the intelligence behind our legal AI platform. You’ll build systems that help lawyers handle complex, real-world legal work, and develop the evaluations and feedback loops that make those systems better over time. This is an applied, product-focused role.
You’ll work across models, prompts, tools, agent infrastructure, and context to improve the quality and reliability of legal work. You’ll work directly with lawyers to understand what good looks like, identify where the system falls short, and turn those insights into measurable improvements. The role has a high degree of ownership. You’ll help define the problems worth solving, make thoughtful decisions with incomplete information, and take work from experimentation through production and iteration.
What You Will Do
- Design, build, and improve AI systems that support complex legal workflows, from understanding a matter through producing and reviewing legal work.
- Develop evaluations that measure real-world outcomes, including the quality, precision, and usefulness of the system’s work, not just performance on isolated benchmarks.
- Improve agent behavior through model selection, prompting, context engineering, tool design, and changes to the surrounding harness.
- Build feedback loops that use lawyer review and final work product to identify failures, prioritize improvements, and measure progress.
- Work directly with lawyers and engineers to translate nuanced legal judgment into testable system behavior.
- Investigate failure modes, design experiments, and make pragmatic tradeoffs between quality, latency, cost, and reliability.
- Own work end to end, from defining an approach and building prototypes to shipping, monitoring, and improving production systems.
- Stay current with advances in models and AI systems, and apply them where they meaningfully improve outcomes.