Lincoln Laboratory engineers, 
admissions officers, and Ivy League counselors. Proud partner of The Haq Foundation.Session 01 Live online
One student, 6,322 schools, and a counselor with 400 of them.
Ninety minutes with the founder of AdmissionsOS and a former UC Berkeley reader on the two halves of this process: the enormous information space, and the human decisions inside it. Live product, real student transcripts, three award descriptions read the way an admissions officer actually reads them.
Format
Live online, 90 minutes
Date
Announced to the list
For
Students, parents, counselors
The problem the talk starts from
College admissions asks a 17 or 18 year old to make a decision that could shape the next decade of their life while they are still going to school every day, taking exams, doing extracurriculars, playing sports, working, spending time with friends and family, and trying to have some version of a normal life.
There are 6,322 post secondary institutions in the United States and another 100,000 institutions around the world. Inside those schools are hundreds of thousands of individual programs and completely different paths through higher education.
The average school counselor is responsible for roughly 400 students. Take everything just described for one student and multiply it by hundreds. There simply are not enough hours in the year.
A counselor may know 50 schools extraordinarily well. Another may know 200. It is arithmetic.
What a school actually costs
Not tuition. Not the number printed at the top of the school's website. Two students at the same university can have radically different real costs.
And the harder problem
What actually matters to me?
What am I unusually interested in, and why do I care?
What environments bring out the best in me?
What constraints have shaped my life?
What makes me different from thousands of other qualified applicants?
Most adults would struggle to answer those questions about themselves. We ask teenagers to answer them while applying to college.
Admissions Eval 101, live in the room
Feiyang reads a file the way a reader does it: in context, at speed, comparatively. The room votes before he says a word.
Applicant 1
Academic
Biostatistics Category 1st Place TNJSF, ISEF Finalist (Top 8 of 375).
This is a science research competition based on a poster presentation.
Applicant 2
Community Service Leadership
The Congressional Gold Medal Award, the highest selective national honor awarded by the United States Congress to youth.
Applicants must complete a minimum of two years of goal-based work, including over 400 hours of community service, 200 hours of physical fitness, 200 hours of personal development, and a five-day expedition. I pursued this distinction by setting long-term goals and exceeding all requirements. I completed 1,000+ hours of service, 2,000+ hours of personal development, and 300+ hours of physical fitness. I additionally designed an expedition focused on Korea, where I conducted cultural research.
Applicant 3
Academic
Invited Presenter, MIT URTC Computational Biology Division.
Requirements: conduct original research suitable for peer review by MIT faculty, graduate students and industry scientists. Selection is highly competitive, with acceptance rates historically in the low-teens. What I did: dedicated 150+ hours in Dr. Akl’s lab on FragPredict, a cancer-focused bioinformatics research project. Analyzed 50,000 compounds, co-built ML model (92.5% accuracy), developed website interface, and published the first ever web-app for cancer drug fragment creation. Co-authored 5-page manuscript and presented a poster to postdoctoral researchers.
Applicant 1 looks impressive in a vacuum, an internationally known science fair. Read again after 2 and 3, and you start to evaluate it differently, more as a missed opportunity. That gap between sounding impressive and being memorable is the lens the rest of the talk depends on.
The two halves, side by side
Sol, on screen
Student
“I need to go to an Ivy League school.”
Sol
Why?
What do you believe an Ivy gives you?
Is that actually unique to those schools?
Would you choose the same environment if nobody knew the school’s name?
What are you optimizing for?
Sol remembers everything, notices patterns, and can say: three months ago you told me this mattered to you. Now you are saying something different. What changed?
Feiyang, in the meeting
Student
“I was the only one coding, so I had to be the only one tracking what's missing and what needs to be done, while everybody else is just waiting for me to tell them... it feels like half the work is on me.”
Feiyang
“It sounds like, as a leader, you had to be both a manager and a micromanager.”
Student
“God, oh my God, YES. that's why I was so... that's why I was salty.”
One word retroactively explained every “salty” she had said, and she claimed it herself. From there she generated her own point for the UC essay.
Run of show
00
Nico introduces himself and Sol, the technology and scale half. Feiyang introduces himself and Your Epic, the admissions insider and human half, landing on the human expert right as the talk turns to understanding the student.
Principle Two halves of one process, named up front.
Hook
Nico frames the core problem: admissions asks a 17 year old to navigate an impossibly large system while a single school counselor juggles roughly 400 students. Feiyang then clears the ground by debunking trending Reddit and TikTok admissions advice, showing how the loudest hacks actually make students blend in.
Principle Technology handles breadth. Humans handle judgment.
101
Feiyang teaches how admissions officers actually read a file, in context, at speed, comparatively, installing the shared lens the rest of the talk depends on. He then walks the room through three real award descriptions and asks which stands out and why, exposing the gap between sounding impressive and being memorable.
Principle A file is never read in a vacuum.
01
Nico demos Sol’s onboarding, how it keeps asking why and pushes back instead of just recording answers. Feiyang shows his read-and-react process through two real student moments, the definition probe and the micromanager recognition, where he reflected a student’s own words back until they clicked.
Principle Personalization is only as good as your understanding of the person.
02
Nico shows Sol surfacing schools and opportunities a student would never have found. Feiyang reframes the spike or hook pressure as a decision problem, helping a student choose confidently among options they already have, using a student who wrote three different drafts to find her topic.
Principle Use technology to expand optionality before using human judgment to narrow it.
03
Nico surfaces several strong-looking schools and Feiyang steps in with the judgment layer, what he would need to know about this specific student before recommending any of them, including the student who chose UT Austin over Berkeley for closeness to home, cost, future NASA aspirations and happiness.
Principle Let technology handle breadth. Let humans handle judgment.
04
Nico shows Sol calculating a student’s true four-year cost after aid, and the pair make the case that financial fit belongs in the equation from day one, not after acceptance.
Principle A school is not truly a fit if the economics do not work.
05
Feiyang debunks how students misuse AI on essays, coining polished abstraction for AI-generated writing that reads better and means less, then shares his VIBE check framework, Voice, Insight, Believability, Engagement, with questions for every student to ask themselves as they revise.
Principle Use AI as a mirror, not a mask.
06
Feiyang runs a live exercise where the audience compares real activity descriptions and feels the impressive-looking ones deflate once broader context lands, driving home why an authentic story beats a college-pleasing one.
Principle The goal is a better-understood student, not an AI-built applicant.
07
The pair show one student against two genuinely different schools, demonstrating how the same authentic person can emphasize different true dimensions without inventing a persona.
Principle Different schools can hear different parts of the same authentic person.
08
Nico shows Sol acting as the operating system for the execution grind, deadlines, forms, supplements, scholarships, so no student loses an opportunity to a forgotten detail. Feiyang: Sol is a god send that I wish every student used and brought to the first meeting.
Principle Technology should make it extremely difficult to lose because you forgot something.
Close
Feiyang leaves the room with the two highest-leverage moves to start now, including the Compound Epic App, to build up raw materials for a future college application no matter what stage of the process you are at.
Principle Start collecting the raw material today.
Who is running it
Momentum Labs, technology and scale
Founder and CEO of Momentum Labs, where I built AdmissionsOS, an end to end admissions platform that uses AI and one of the largest admissions databases in the world to make the entire college application process simple. The platform has reached 50,000+ students, and I have personally coached 200+ students through admissions, with students earning admission to every top 100 university in America.
Previously, I spent two years at MIT Lincoln Laboratory building machine learning systems for Air Force platforms, co-invented Vein Guard which is a potential cure to fatalities caused by deep vein thrombosis, and founded Project Snake Eyes, a special forces helmet designed to detect the location of shooters the instant they fire.
I was also a quantitative researcher at Cinctive Capital, working on a macro trading desk managing more than $2 billion of options exposure, and later founded Gradient Trading, an algorithmic trading hedge fund that raised $3.4 million. I operate research infrastructure across 350 million+ academic papers and run a technical consulting company where I solve technical problems for startups and venture firms, from cybersecurity to nuclear fission diligence.
I also co-founded the Haq Foundation alongside Zar the Star, TheWizardLiz, and Chef Tova, who collectively reach more than 23 million followers, with Momentum Labs providing the education infrastructure as we build toward helping 20,000 Afghan girls continue school, apply to universities around the world, secure funding, and actually evacuate Afghanistan and get there.
Your Epic, admissions insider and authenticity
Founder of Your Epic Consulting, where I help students get into college as their most authentic selves, not the polished, 'college-pleasing' version they think admissions officers want to see.
Before founding Your Epic, I spent 4+ years inside UC Berkeley doing admissions work, reviewing 2k+ applications, evaluating well over 10k+ UC PIQs. While my former students have earned admission to top schools including UCLA, Berkeley, Stanford, UT Austin, and Johns Hopkins, more importantly many realize through this process, in my former student's own words, “that working with me through this process has strengthened their self-confidence”
I'm a proud Bruin and stay close to the UCLA family. I am a former VP of the NYC Tri-State Bruins, ex-President of the Sacramento Bruins, and a 2025 Bruin Business 100 honoree. I'm especially proud of guiding double-Bruin families through the complex admissions process, and of mentoring current UCLA students and recent grads through their academic, career and life transitions.
I also co-founded Sonder, one of the first truly student-driven college admissions communities, alongside one of my current students who is a high school junior going through the process in real time. Sonder is built on the reverse principle of reddit fear-mongering, where students are reminded that ‘no one can do the work for you, but you don't have to do it alone.' Students are in an environment where they learn as much from each other as they do from me.
Three ideas to leave with
01
Start with the student, not the schools.
02
Use AI as a mirror, not a mask.
03
Let technology handle breadth. Let humans handle judgment.
