Outcome-driven college recommendations.

CollegeAI is a personalized, AI-assisted college recommendation and resource platform for high-school students. The CollegeAI engine analyzes a student profile and outputs scores and personalized reasons why the student might like or dislike each college in the United States.

CollegeAI fixes college search. Most college search websites allow students to filter by broad categories, such as cost and school size, but college research is a delicate balance of hundreds of factors, including merit-aid and financial-aid adjusted cost, rankings, community appeal, program offerings, campus aesthetics, dating culture and much more.

CollegeAI is the only service that leverages artificial intelligence and historical student outcome data to match students to colleges based on extensive student profiling. Today, over 30,000 students have used CollegeAI to enhance their college search and discovery process.

CollegeAI began as a research project at Rensselaer Polytechnic Institute in September of 2016. Our goal was to predict where a high school student would end up going to college based on a set of easy to answer questions that assess the student's personality, interests and academic performance. We then launched CollegeAI.com, an online platform that allows students to answer a set of questions then receive college recommendations. We supply the student with personalized reasons why we recommended each college (we call them the "pros and cons"). Students can explore colleges, create an application list and compare colleges based on a variety of factors. When the student has been accepted into colleges, they can also get a final choice recommendation.

We're making college help more affordable, the average advising package for independent education counselors is $4,400 [0], and about 160,000 students utilize independent education counselors [1]. Our platform is totally free, so anyone can get access to basic college counseling.

Our team is made up of 2 full time software developers, a Harvard-trained designer, and 2 professors from Rensselaer Polytechnic Institute who study machine learning, artificial intelligence and social choice and several key advisors from the education space.

Before we raised money, we were just 2 students working from an apartment. When our lease ran out, lived and worked inside a single tiny room of a fraternity house for two months (it was the cheapest available housing). We then finally got some initial funding and were able to move to Cambridge, MA. Now we work from inside the LearnLaunch Boston campus with a ton of other early stage edtech startups.

Check out CollegeAI yourself by pretending to be a student on collegeai.com, or hit "Images" on the sidebar to see screenshots of the product.

[0] Mark Sklarow.Trends in independent educational consulting.Technical report,Independent Education Counselor Association,2015.
[1] Sarah Brachman. National study shows dramatic increase in hiring private college counselors. Technical report, Independent Education Counselor Association,2009.

The Problem CollegeAI Solves

The root of problem is the difficulty of college analysis. First generation students often have no idea what they should do to prepare for college and what colleges they should be looking at. This problem is exacerbated if they're in a underserved community where the average student to counselor ratio is 500:1 (across the United States). Counselors at schools with poor student counselor ratios aren't really focused on effective college counseling- they need to ensure student wellfare, help with class scheduling, and numerous other responsibilities. CollegeAI will assist in the effort to make education more equitable by providing every student with effective college counseling.

For affluent families, CollegeAI can prevent wasted tuition by enabling students to make better college decisions, reduce the number of college visits a student must do, reduce college anxiety and uncertainty for parents and students and help students make effective application decisions.

CollegeAI's Innovation

We match students with colleges based on historical outcome data, meaning we look at where previous students who went through our system went to school, what they liked/disliked, then use that to help future students who are similar. Other services just filter colleges, we're more like a matchmaking service.

Our artificial intelligence (internally known as The Match Maker) uses research performed under Malik Magdon-Ismail at Rensselaer Polytechnic Institute, and leverages the latest developments in machine learning, social choice theory, recommendation systems and over 330,000 student data points.