Skip to main content

Stevens Institute of TechnologyCollege of Professional Education

PULSE.AI Lab

Applied AI for real workflows, built with our students.

Students learn alongside experts from both research and industry, turning real problems into working AI.

Why the lab exists

We build AI capability the Stevens way, by doing.

P.U.L.S.E. stands for Product, User experience, Learning systems and Engineering.

  1. Learn alongside experts

    Students work with experts from both research and industry.

  2. Optimize real workflows

    Teams improve the work people already do and create new AI solutions.

  3. Keep the pipeline moving

    Every project trains the next builders and keeps the talent pipeline moving.

What the lab does

Put AI to work at Stevens

We design and run AI workflows and products across the college and with partners, from operations to teaching to student services. If it gives people hours back every week, it belongs here.

Projects at every stage, many not public yet

  1. Live in daily use
  2. In pilot
  3. Early development
  4. In stealth
See selected projects

Give students the real thing

Students work on live problems with faculty, researchers and industry experts, with real users, deadlines and data rules. Not a case study. The work ships.

Keep the pipeline full

Every semester brings new problems in and new talent through. Projects outlast a cohort, so each team inherits working systems and leaves them better.

How a project moves through the lab

Six steps, in a loop. What we ship raises the next round of questions, and the people who built it help the ones who come next.

  1. 01

    A problem comes in

    From a partner company, a faculty researcher or a CPE team. We ask who feels the pain, how often, and what better would look like.

    Leaves behind: Problem brief

  2. 02

    Frame it with experts

    Faculty and practitioners help define the data, the constraints and the measure of success before anyone writes code.

    Leaves behind: Scope and success measure

  3. 03

    Form the team

    Students are matched by skill and interest, with a lab lead and a domain mentor. Everyone knows what they own.

    Leaves behind: Student team and mentors

  4. 04

    Build in short sprints

    Working software every week, demoed to the people who asked for it. Feedback lands while it is still cheap to change.

    Leaves behind: Weekly working demos

  5. 05

    Prove it before it scales

    We test with real users and check accuracy, accessibility, privacy and cost. If it is not better than the old way, it does not ship.

    Leaves behind: Evaluation and sign-off

  6. 06

    Ship it and pass it on

    The system goes into daily use with an owner and a way to measure it. Students leave with shipped work, and the next cohort picks up the roadmap.

    Leaves behind: A system in use, and the people who built it

A sample of the work

A few projects already in use at Stevens. Many more are in pilot, early development or stealth, so they are not shown here.

Iris

Executive intelligence for the college.

Live enrollment and pipeline insight for CPE leadership, with Navs, an assistant that answers questions about the data in plain language.

The question behind it
Can leaders ask their data a question and trust the answer?
Used by
CPE leadership

Remedy

Accessible course materials, automatically.

Checks PowerPoint, Word and PDF files against WCAG 2.1 and Section 508, fixes what can be fixed safely, writes real alt text, and reports what still needs a person.

The question behind it
How much accessibility work can a machine do safely, and where must a person decide?
Used by
Course production teams

EduForge

The whole life of a course, in one place.

Faculty draft course outlines with AI help, the studio runs production and quality checks, and vendors build in Canvas, with every step tracked.

The question behind it
What changes about making a course when AI is part of the team?
Used by
Faculty, studio and vendor teams

Steve

An AI career agent for Stevens students.

Measures readiness for a target role, verifies skills with job-realistic assessments, rehearses interviews round by round, keeps the resume current, and brings in a Stevens advisor when a person is better.

The question behind it
Can an AI agent judge job readiness honestly, and know when to hand off?
Used by
Students and career advisors

Slide Studio

Any deck, rebuilt on brand.

Turns rough presentations into Stevens-branded decks, then reviews every slide against the source so nothing is lost in the redesign.

The question behind it
Can design judgment be automated without losing the message?
Used by
Faculty and staff
See selected projects in depth

Questions on the board

A sample of what teams have taken on. New ones arrive every semester.

  • Which admitted students need a nudge this week?
  • Can every course file pass accessibility before launch?
  • What should a student learn next to land the role they want?
  • How long does it really take to build a course?
  • Can a rough deck be on brand by morning?
  • Which skills are employers asking for next?
  • Where does a learner get stuck, and why?
  • Can an assistant answer from approved program data only?

Bring us a problem that sits between learning, technology and work.

Partners and researchers

Tell us the problem, who it affects and what better looks like. We will tell you honestly whether the lab is the right fit.

Bring us a problem

Stevens students

Want to work on a live project? Tell us what you study and what you want to build.

Ask about joining

Signals become systems. Systems become learning. Learning becomes capability.