Back to index
Other9 courses (36 units), about two years and a quarter at one course per quarter, three-year maximum·$1,155 per unit, $41,580 total tuition, plus a $135 to $155 application fee

UCLA MSOL: Master of Science in Engineering Online, Data Science Engineering

3.9

A real UCLA engineering master's you can finish without ever attending a live session, built from the same graduate ML courses the campus students take. Fully asynchronous, mid-range pricing and a diploma that does not say online. The catches are a GRE requirement that most competitors dropped years ago, a degree title that says Engineering rather than Computer Science or Data Science, and an engineering, maths or physics bachelor's as the entry ticket.

What We Liked

  • Course pool includes CS 260 Machine Learning Algorithms, 260C Deep Learning, 260D Large-Scale ML, 260R Reinforcement Learning, 263 NLP and 260F Foundation Models
  • Fully asynchronous, so it works across time zones and shift patterns
  • $41,580 total is about half of Columbia or USC online and under two thirds of Stanford HCP
  • Diploma reads Master of Science in Engineering with no online designation, plus a Certificate of Specialization in Data Science
  • Ranked number one online engineering master's by U.S. News in 2026, the eighth time in twelve years
  • Capstone project option under a UCLA faculty member as an alternative to written comprehensive exam questions

What Could Be Better

  • GRE still required, waived only for highly competitive portfolios on a case-by-case basis
  • Degree title is MS in Engineering, not Computer Science or Data Science, which needs explaining on a resume
  • Requires a bachelor's in engineering, mathematics or physics, so career changers from other fields are out
  • Only two intakes a year and a hard three-academic-year limit
  • Still four to six times the price of Georgia Tech OMSCS or UT Austin
  • The program's own pages quote two different tuition totals, $39,600 and $41,580, which is sloppy for a $40,000 purchase

Detailed review

UCLA's MSOL is the Samueli School of Engineering's online master's, and the Data Science Engineering specialization is the one that overlaps most with what this site covers. It is a 36-unit degree, nine courses, with five drawn from a core list and four electives. At least five courses must be at the graduate 200 level, excluding the capstone. The core list is where the program earns its keep. It includes COM SCI 260 Machine Learning Algorithms, 260B Algorithmic Machine Learning, 260C Deep Learning, 260D Large-Scale Machine Learning, 260R Reinforcement Learning, 262A Learning and Reasoning with Bayesian Networks, 263 Natural Language Processing and the newer 260F Foundation Models, alongside data infrastructure courses such as EC ENGR 219 Large-Scale Data Mining, COM SCI 143 Data Management Systems, 245 Big Data Analytics and 232E Large-Scale Social and Complex Networks.

These are the department's graduate courses, not a separate online syllabus, and you can see the depth from the numbers alone. Every student has to satisfy a comprehensive exam requirement, and UCLA gives you two routes. One is the ENGR 299 capstone project, four units of supervised work with a UCLA Engineering faculty member, available once you have five 200-level courses behind you. The other is to answer three designated written exam questions from three different graduate courses in your specialization, sat alongside those courses' finals. I like that the project route exists, because a supervised piece of work is worth more in an interview than a transcript line. The program is fully asynchronous, which UCLA states plainly.

There are no live sessions to attend, and the program says it can be completed anywhere with an internet connection. At one course per quarter, including two summer sessions, most students finish in two years and a quarter. Doubling up gets you down to a year and a half to two years. The maximum is three academic years, nine quarters, excluding summers. On money, the program page lists $1,155 per unit and $41,580 in total tuition.

The FAQ on the same site still quotes $39,600, which I take to be last year's figure, and I would plan on the higher number. The application fee is $135 for U.S. citizens and permanent residents and $155 for everyone else. That lands UCLA in the middle of the online master's market: around half of Columbia CVN or USC DEN, roughly four times UT Austin's online MSAI and about six times Georgia Tech's OMSCS. Admissions are the awkward part.

You need a bachelor's from a regionally accredited institution in an engineering discipline, mathematics or physics, a GPA of at least 3.0 with some flexibility, and proficiency in programming including Python. You also need three recommendation letters, a personal statement and a statement of purpose. And you need the GRE. UCLA says it may waive it for applicants with highly competitive holistic portfolios, decided case by case, but as a default requirement it is out of step with Georgia Tech, UT Austin, Columbia and USC, all of which have dropped it.

The next start is March 29, 2027 with a December 1, 2026 deadline, and there are only two intakes a year. The other thing to think through is the degree name. The diploma says Master of Science in Engineering, with a separate Certificate of Specialization in Data Science. There is no mention of online anywhere on the diploma or transcript, which is good, but there is also no mention of computer science or data science on the diploma itself. For most hiring managers a UCLA engineering master's with ML coursework is plenty.

For automated resume filters looking for an MS in Computer Science it is a small, real disadvantage. My honest take is that UCLA has built one of the more sensible online programs at a top-tier school. The courses are real, the format respects working adults, the price is high but not absurd, and the brand is strong on the West Coast and beyond. If I had an engineering bachelor's, a tuition benefit covering half, and no appetite for live classes, this would be near the top of my list. The GRE and the entry restrictions are the reasons it is not a default recommendation.

[ final ]

The verdict.

The best fit of the big-name online master's for a working engineer who wants a strong-brand ML degree without live class times and without an Ivy League bill. If you cannot or will not sit the GRE, or if the degree name matters more to you than the course content, look at UT Austin or Georgia Tech instead.