Welcome back, Alex

No research experience

USACO Silver and two shipped apps demonstrate self-directed ability, but neither involves a faculty collaborator or published output. MIT and Stanford CS admits almost uniformly show at least one externally mentored project — a professor lab, a named publication, or a recognized program. Without this, the profile reads as technically capable but unverified at an academic level.

→ Students in this position take different paths — some cold-email professors with a specific research question framed around a project they've already built; others apply to structured programs like MIT PRIMES, RSI, or SIMR where the connection is made for them. The right path depends on timeline, geographic access, and whether the student has a research question they can articulate.

SAT ceiling for reach schools

SAT 1520 is below MIT's reported 50th-percentile of ~1580 and Stanford's of ~1570. The Math section (770) has less headroom than Reading & Writing (750) based on the most recent attempt — and R&W is the section most improvable with targeted practice.

→ Some students in this position retake once to close the gap; others decide the profile is strong enough elsewhere and invest the preparation time in a project or competition instead. The decision depends on whether a 40+ point improvement is realistic given prior score trajectories.

Study Planner lacks outcome data

The AI Flashcard Generator entry cites 300+ active users — a concrete metric that makes the claim verifiable. The Study Planner entry, listed separately, does not include any user count, retention figure, or launch date. Reviewers reading both entries back-to-back will notice the gap in rigor.

→ Students close this by adding a specific metric — DAU, weekly actives, a notable user story, or a growth figure. The format matters less than having at least one external reference point that a reader can verify or visualize.