Original post on bemyapp.com by Tony Alioto

April 14, 2016
Reading time: 4 minutes

Dan Rubins

In this, the information age, we are continually searching for new avenues where technology may improve and simplify our lives – and this is especially true in regards to artificial intelligence, and the role we hope it may one day play.

Dan Rubins is the CEO of Legal Robot. His company uses machine learning techniques to analyze legal documents, and check for errors that are often left unnoticed by the human eye. Their algorithms give users as much possible information by comparing thousands of other legal documents, and issuing any potential risks or issues.

Legal Robot is a member of Hacker Unit’s inaugural class of startups, and is garnering lots of attention, both in the legal and tech worlds. Check out their FAQ page, there’s lots of great material both on deep learning and AI, and the unique story of their company.

Dan sat down with us at BeMyApp to answer a few questions about Legal Robot, his business model, and who’s inspired up until this point in his life…

Q: What problem do you solve?

A: Legal language is hard to understand and causes a lot of unnecessary friction in the global economy. We provide automatic legal document review using artificial intelligence. We can summarize legal language, break it into manageable chunks, and present it more visually.

Q: Who is your target consumer?

A: Everyone is a consumer of legal language, whether they know it or not. So, we want to help everyone understand what they are signing. More targeted uses of our technology are important for businesses and law firms, especially in due diligence, auditing, and contract review.

Q: What is your business model?

A: Sporadic users, like consumers, can pay $15 per document – certainly cheaper than $400/hr for an attorney. We have usage-based subscription tiers for higher volume users like law firms and enterprises.

Q: What challenges have you faced, tech or otherwise?

A: The technical challenges are significant. Working with legal language is difficult because most natural language processing tools are trained on plain language, not highly structured legal language. However, the biggest challenge has been finding sufficient data to feed our machine learning models.

Q: Who inspires you?

A: My co-founder and wife, Megan Satterfield-Smith has been my sounding board and voice of reason for the last 15 years. Startups are hard and it is important to have a strong support system. One of the most common causes of startup demise is founder disagreements; too often, founders ignore the value of a track record.

Q: Where do you see your company in 5 years?

A: We will be working in mediums other than legal contracts as well as in multiple languages. We want to play a huge role in legal services innovation as well as improving Access to Justice. Our technology will define new ways for people to interact with the law.

Q: What have you learned since beginning this startup journey?

A: Solving a real problem for a large number of people is far more powerful (and valuable) than a nicely polished but niche product.

Q: Most memorable story from the startup process?

A: It was really exciting the first time a deep learning model told me something I didn’t know. When you are building a model, you’re very often just looking at numeric statistics like error rates, AUCs, etc. It’s very dry and doesn’t feel very real. So, I glanced at some text, made my own judgment about it and then got a different (and more specific) answer from my model. Of course I thought this was an error at first, but upon close reading of the text, it was really interesting to see that the prediction was correct and I was wrong.

Q: In 3 bullet points, why a VC should invest in you?


  • Market. Legal Services is huge and incredibly regressive, which makes it ripe for change.
  • Technology. We’re building on the recent advances in machine learning and natural language processing, but creating new approaches tailored for highly structured legal language.
  • We’re solving a real problem. The potential impact of unlocking all of the information buried in legal texts is staggering.

Q: Your favorite city or place in the world?

A: I’ve been lucky enough to travel to many amazing places but California is still my favorite.

Q: Favorite 3 Movies?

A: Inception, Lethal Weapon, Monty Python and the Holy Grail

Q: Books that are important to you?

A: Zero to One by Peter Thiel inspired me to solve much bigger and more important problems than I previously considered. Also, The Hard Thing About Hard Things by Ben Horowitz is a great inside look at the way companies actually operate and an interesting narrative about key decisions as a CEO.

Q: What superpower do you wish you could have?

A: Time travel.

Q: What’s your mantra?

A: Action conquers fear.