Resume Teardown #77: CS Undergrad, Big-Tech Intern, 58%, An Engineer's Resume Applying for PM
This is part of our Resume Teardown series where we score real PM resumes (anonymized) and break down what the evaluation found.
TL;DR: A computer science undergrad at a top engineering college, with a software engineering internship at a major professional networking platform, scored 58% for PM and APM roles. The technical execution is genuinely strong. The resume reads as an engineering resume, with no evidence of how this candidate makes product decisions.
The Resume
Background: Third-year B.Tech Computer Science student at a top engineering college with a strong academic record. Software Engineering Intern at a major professional networking platform, working on recruiter-to-ATS activation workflows. Two notable projects: an academic portfolio tool with authentication, and an AI agent for e-commerce chargeback arbitration using multimodal reasoning and computer vision. Strong competitive programming ratings and selective mentorship and accelerator program selections.
What looked good on the surface:
- A big-tech software internship is a strong brand signal and real production exposure
- The AI project has substance: it names specific techniques (multimodal reasoning, OCR, pixel-level error-level analysis) tied to a concrete fraud-detection use case, not generic AI name-dropping
- High execution rigor evidenced by competitive programming ratings, hackathon placements, and selective program admissions
- Clean technical skills coverage across languages, web stack, databases, and ML
Score: 58%
Why a Strong Student Resume Scored 58% for PM
This resume would score well for a software engineering internship. It scored 58% for PM roles because it is a software engineering resume. That is not a criticism of the candidate. It is a framing mismatch.
The scoring is weighted heavily toward skills (60%) at the undergrad level, and the gap it found is specific: there is no evidence of product craft. No user research, no requirements definition, no prioritization, no experimentation, no launch measurement. Everything on the resume describes what was built and how, not why it was built or what changed because of it.
The hiring manager verdict captures the exact decision a recruiter would make:
The engineering potential is clear, but I would pass for most APM screens right now because I cannot tell how this person makes product decisions, learns from users, or measures outcomes.
The good news for a student: this is the most fixable gap in the entire teardown series. The raw material is here. It needs reframing, not more experience.
The Internship Bullets: Real Work, Engineering Framing
The internship is the strongest asset on the resume, and it is described entirely in engineering terms.
"Automated the activation workflow between LinkedIn Recruiter and partner Applicant Tracking Systems (ATS), eliminating manual configuration steps."
The feature is clear. The value is not. For a PM resume, the question is: how many configuration steps were eliminated, how much activation time was saved, how much support effort was avoided, or how many partners the automation now covers? There is a product outcome hiding behind this bullet. It is not stated.
"Engineered a full-stack preference-override mechanism (TypeScript, Java, MySQL), exposing RESTful endpoints to persist user state and dynamically bypass default external API payloads on subsequent requests."
This is technically credible and reads like a pull request description. For a PM role, the implementation detail should move to the background and the product rationale should move to the front: what user or operational problem forced an override mechanism, and what was the trade-off in bypassing default payloads?
The fix pattern:
Before: "Engineered a full-stack preference-override mechanism (TypeScript, Java, MySQL), exposing RESTful endpoints to persist user state and dynamically bypass default external API payloads."
After: "Shipped a preference-override system so recruiters could keep custom ATS configurations across sessions, resolving a repeat-setup pain point. Chose to persist user state server-side over client caching to keep behavior consistent across partner integrations."
Same work. The after version shows a user problem and a decision. That is the difference between an engineering bullet and a PM bullet.
The AI Project: Substance Without Validation
The chargeback arbitration project is the most PM-promising thing on the resume, because it starts from an operational problem rather than a technology.
"Built an AI agent to automate e-commerce chargeback arbitration, ROI calculations, and bank-compliant PDF generation."
"Integrated Gemini's multimodal reasoning with PaddleOCR and pixel-level ELA for zero-shot visual fraud detection."
The technical ambition is real and specific. What is missing is the product-quality thinking: how good is it, and when does it fail? For any AI product role, that is the first question an interviewer asks.
The resume names the components but not what each contributes, how conflicting signals are resolved, or what happens when the model is uncertain. Add the evaluation approach: representative test cases, a precision or false-positive figure, a processing-time comparison, and the human-review fallback for low-confidence cases. AI PM work is judged on exactly those decisions, not on the model list.
Because this resume triggered the AI PM signal in scoring, this project is the candidate's strongest lever. Making its decision layer visible would move multiple dimensions at once.
The Missing Summary
There is no professional summary, and for a career-starter pivoting from engineering toward product, that is the single highest-impact omission.
A recruiter reading this resume has no statement of product direction. They see a strong engineering student and reasonably assume the candidate wants a software engineering role. Nothing redirects that assumption.
A two-line summary that connects the engineering background to the product problems this candidate wants to own would reframe how every bullet below it reads. It is the difference between "a CS student who also applied to PM" and "an engineer who thinks about products." The experience does not change. The lens the recruiter uses to read it does.
Dimension Scores
Skills and Tools: 57%
The lowest dimension, and the heaviest weighted at this level. Strong technical execution: TypeScript, Java, MySQL, REST APIs, React, Node, and production full-stack work. The gap is zero evidence of PM craft: user research, requirements, prioritization, analytics, experimentation, or launch measurement. The fix is adding PM-relevant framing and artifacts from the internship and projects, not learning a new tool.
Leadership and Impact: 50%
The resume does not show how the candidate identified a user need, prioritized scope, or influenced collaborators on a product decision. The internship and the AI project both contain the raw material for this. It is a matter of surfacing the discovery and trade-off steps that already happened.
Experience and Background: 63%
The internship brand and the academic and competitive record are genuine strengths. The gap is that the resume reads as an engineering student resume with no product direction stated. A PM-oriented summary addresses this directly.
Domain Expertise: 70%
The highest dimension, somewhat surprisingly. The signals span recruiting technology, e-commerce fraud, and academic tooling, so no single niche is yet the candidate's clear hiring domain. That is normal for an undergrad. Pick the area with the strongest user-problem understanding (recruiting tech or fraud/payments both have depth here) and keep building evidence there.
ATS Readiness: 42%
This is the lowest ATS score we have published in a while, and it is almost entirely a formatting problem rather than a content problem.
Pass: Standard section headers (Education, Experience, Technical Skills, Projects, Achievements), consistent date formats.
Warning (spelling): Words are running together in the extracted text, such as "Automatedthe" and "afull-stack." This is not a typing error. It signals that the document's spacing is not surviving text extraction, which usually points to a template built with text boxes, tables, or non-standard spacing.
Warning (acronyms): ELA, JWT, CNNs, RNNs, DBMS, and OOP are not expanded. Spell out the less-universal ones on first use.
Fail (keywords): Only three PM keywords were detected across the entire resume (requirements, user, impact). Missing: roadmap, stakeholder, strategy, metrics, prioritization, cross-functional, discovery, launch, user research, data-driven, outcome, adoption, experimentation, iteration, trade-off, backlog. For a PM application, this is the core signal, and it is largely absent from the experience and project bullets.
The root cause of the 42% is the template. Stray symbols near the name, fragmented table text in the education section, and merged words all point to a layout that parses poorly. Switching to a clean single-column template with plain-text headers and standard hyperlinks would lift this score substantially, and the keyword additions would come naturally once the bullets are reframed in product language.
Key Takeaways
1. An engineering resume and a PM resume describe the same work differently. The internship work is real. Described as implementation ("engineered a mechanism exposing RESTful endpoints"), it signals engineer. Described as a decision ("chose to persist state server-side to keep behavior consistent across partners"), it signals PM. Reframe, do not rewrite.
2. For a transition, the summary is not optional. Without a stated product direction, a recruiter defaults to reading you as what your title says. Two lines connecting your engineering background to the product problems you want to own changes how every bullet below is interpreted.
3. AI projects are judged on validation, not on the model list. Naming the components (multimodal reasoning, OCR, error-level analysis) shows technical range. Showing precision, false-positive rate, and the human-review fallback shows product judgment. The second is what AI PM interviewers screen for.
4. Add the "why" and the "what changed" to every bullet. "Automated the activation workflow" becomes a PM bullet the moment you add the baseline and the outcome: steps eliminated, time saved, partners covered. The outcome data exists. Put it on the page.
5. A 42% ATS score is usually a template problem, not a content problem. Merged words and stray symbols in extraction mean the layout is breaking. A single-column, plain-text-header template fixes this in one pass and protects every future application.
The Pattern
This is a "right work, wrong framing" resume from a strong engineering student targeting product roles.
The candidate has done genuine product-adjacent work: owned a defined integration problem at a major platform, and built an AI agent around a real operational pain point. But all of it is described in engineering language, so a recruiter mis-categorizes the candidate as an engineer who happened to apply for PM. The fix is not more experience or more projects. It is reframing the existing work around user problems, decisions, trade-offs, and outcomes, and stating a clear product direction up top. This is the most common and most fixable student-to-PM pattern we see.
Score your own resume to see how your product manager resume performs across all four dimensions.