R.V. showed real technical growth without a research publication.
He had three years of useful software work, but no publications, a general job title, and code he couldn’t share. We helped him show his work through evidence anyone could inspect and compare programs against his family’s budget.
The starting point
A software analyst at a logistics company in the Philippines, applying after three years of work
Field of study
Computer systems, machine learning infrastructure, and reliable software
An anonymized student journey, shared with initials.
AI-generated illustration, not a photograph of R.V.
Useful software work, hidden behind a general title.
R.V. maintained internal software used by warehouse teams and had taken on monitoring and data-cleaning work. Confidentiality kept him from sharing the code. With no publications, he first described routine automation as artificial-intelligence deployment, and his goals moved between research, product management, data science, and entrepreneurship.
HOW WE HELPED
We helped him describe his tools by what they actually did, show public evidence of his method, and narrow his goal to reliable computing systems.
He kept warehouse software running.
He maintained internal tools used by warehouse teams and responded to operational incidents.
He took on monitoring and data cleaning.
His responsibility grew to include monitoring and data-cleaning workflows.
His best work was confidential.
He couldn’t share proprietary code, so he needed other ways to show how he worked.
THE PUBLIC PROJECT
A small public project was stronger than an inflated AI claim.
To show his method without breaking confidentiality, R.V. built a small public project that reproduced a scheduling problem using sample data he created.
THE FIRST DRAFTS
Routine automation was called AI.
A technical reader could see the gap between the label and the work.
→
THE PROJECT
A scheduling problem, rebuilt with sample data.
It showed his method and documentation from start to finish, without claiming innovation.
→
THE FOCUS
Reliable computing systems.
His statement narrowed to the operational limits he had actually met at work.
LETRADO’S ROLE
We tested the logic of his goals and asked where evidence was missing. He selected the examples, wrote the code and application materials, and verified every figure.
HOW WE WORKED TOGETHER
Preparation made his work visible and his decision affordable.
R.V. contributed regularly to his family’s expenses. The most prestigious offer could also mean the largest withdrawal from savings, so cost was part of fit from the start.
An inventory of his evidence.
He listed university coursework, public code, systems incidents, team responsibilities, and self-directed study.
Specific without breaking confidentiality
Confidential projects were described at an appropriate level, with no invented performance figures. His employer wasn’t asked to disclose anything.
Recommenders got a factual brief.
They received a list of projects they had directly observed, not drafted praise to sign.
Growth, not only reliability
His recommendations first focused on how dependable he was. The brief helped his recommenders describe his technical growth too.
Programs were compared with the full cost.
He weighed degree format, research or professional focus, capstone work, recruiting, and cost.
Lost salary counted too
The comparison included tuition, living costs, lost salary, program length, and the uncertainty of finding work abroad afterward.
Cornell’s MEng fit the work he wanted and a cost his family could manage.
R.V. was also admitted to Stanford’s MS in Computer Science. He chose Cornell’s MEng in Computer Science after comparing professional focus, program length, family funding, and the cost of leaving work.
A professional degree for practical systems work.Its professional focus and structure matched the reliable-systems work he wanted to do next.
A realistic total cost.Stanford’s offer was attractive, but its total cost and degree structure led him elsewhere.
