Detecting Soft Skills in ML Engineering Roles CVs

2026-08-10

AI Summary

This study addresses the limitations of existing resume mining approaches, which rely heavily on keyword matching and struggle to identify narrative expressions of soft skills, while lacking empirical analysis of discrepancies between candidate self-presentation and employer expectations. The authors propose a large language model–based extraction pipeline to detect both explicit and implicit soft skills from a balanced sample of 300 resumes, and operationalize employer claims from job-related literature into 13 falsifiable hypotheses subjected to statistical testing. The work’s novelty lies in its first effective extraction of narrative soft skills and the establishment of a verifiable comparative framework between candidate and employer perspectives. Through human-annotated validation and effect size analyses controlling for family-wise error, 11 of the 13 hypotheses are supported: candidates express soft skills narratively at a ratio of approximately 3:1; leadership is nearly three times more likely to appear in senior-level resumes; and software engineers mention leadership at only half the rate of their peers.

Abstract

Soft skills shape collaboration among ML engineers, data scientists, and software engineers building ML-enabled systems, yet what we know about them comes almost entirely from the demand side. Job advertisements, surveys, and hiring manager interviews capture what employers ask for. How candidates themselves articulate these competencies has not been studied, and existing CV-mining work is both keyword-based, so it cannot see skills conveyed through narrative, and descriptive, reporting frequency rankings without testing whether group differences exceed sampling variation. We close both gaps. Using a balanced corpus of 300 curated CVs spanning the three roles, we extract explicitly listed and implicitly narrated soft skills with an LLM-based pipeline validated against a human-annotated ground truth, a distinction that existing extractors were not designed to make. We then convert the demand-side literature's claims into 13 falsifiable hypotheses about role signatures, seniority progression, and disclosure style, and test them with effect sizes under family-wise error control, so that candidate-side data can corroborate or contradict the demand-side account rather than merely illustrate it. Eleven hypotheses are supported, one partially, and one refuted. Candidates disclose soft skills through narrative rather than keyword lists by roughly three to one, and most so for the competencies employers value most: leadership, coordination, and mentoring (88-96% narrative). Seniority nearly triples the odds of articulating leadership. That competency, assumed universal in prior work, is articulated by software engineers at half the rate of their peers. Technical candidates do articulate soft skills, but a keyword-based screening systematically misses them.

Problem

• soft skills
• CV analysis
• machine learning engineering
• narrative expression
• role differences

Innovation

• LLM-based extraction
• narrative soft skills
• CV mining
• hypothesis testing
• family-wise error control