Inspiration
Breaking into tech as a student or junior developer can be frustrating.
A CV can say someone knows Flutter, Python, React, APIs, or cloud technologies—but a recruiter still has to ask:
“What can this person actually prove?”
We built SkillProof around that question.
We wanted to move the focus from self-reported skills to observable evidence from real work. A developer's projects, repositories, deployments, documentation, and engineering practices can tell a much more useful story than a list of technologies on a résumé.
The idea was simple: turn projects into proof.
What We Built
SkillProof analyzes publicly available GitHub project evidence and turns it into an understandable skill profile.
Instead of only saying:
Flutter — Yes
SkillProof asks:
What evidence supports that skill?
The product analyzes project signals such as repositories, languages, commits, documentation, testing indicators, CI/CD configuration, and other observable engineering activity.
It then organizes the evidence into three outcomes:
Skills with evidence
What the developer has already demonstrated.
Evidence gaps
Where the developer claims or appears interested in a skill but has limited observable proof.
Next Proof
A practical project or improvement that can generate stronger evidence for the missing area.
The goal is to create a continuous loop:
Build → Prove → Identify gaps → Build again → Prove again
How We Built It
We designed SkillProof as a lightweight web application that can analyze a public GitHub profile without requiring the user to configure a complicated developer environment.
The MVP includes:
- GitHub profile import
- Public repository analysis
- Technology and project evidence extraction
- Skill-strength visualization
- Evidence-gap detection
- Repository-level evidence
- Recommended next projects
- Demo mode for reliable presentations
- Responsive interface and exportable reporting
The scoring layer is deliberately transparent. We did not want a mysterious AI-generated number that users cannot challenge or understand.
Each skill signal is connected to observable evidence, making it possible to ask:
“Why does SkillProof think I have this skill?”
and receive an understandable answer.
What We Learned
Our biggest lesson was that a skill is more valuable when it is demonstrated than when it is declared.
We also learned that junior developers often do not have an evidence problem because they are incapable; they have an evidence problem because nobody has helped them understand which artifacts actually demonstrate competence.
That changed our product direction.
Instead of building another résumé generator, we focused on helping users understand the relationship between:
what they build → what it demonstrates → what is missing → what to build next.
We also learned that transparency matters. A score without evidence can create false confidence, so the product emphasizes the underlying repository signals rather than treating the score as an absolute measure of ability.
Challenges
The biggest challenge was distinguishing activity from meaningful evidence.
A repository with hundreds of commits does not automatically prove strong engineering ability. Likewise, simply using a programming language in one file does not demonstrate deep expertise.
We therefore designed SkillProof around multiple signals rather than a single metric.
Another challenge was working within the limits of public GitHub data. Public repositories do not reveal everything about a developer's actual abilities, especially private work, teamwork, interviews, architecture decisions, and communication.
That led us to an important product principle:
SkillProof measures evidence, not human potential.
The current system should support a developer or reviewer, not replace a hiring decision.
Why It Matters
Students and junior developers often face the same problem: they are asked to demonstrate experience before they have had many opportunities to gain formal experience.
SkillProof gives them another path.
Instead of saying:
“I know Python.”
they can progressively build evidence that shows:
“Here are the systems I built, here is how they work, here are the engineering practices I used, and here is the evidence behind the skill.”
Our long-term vision is to make SkillProof a continuous proof-of-skill layer for developers—connecting portfolios, GitHub activity, deployed applications, technical assessments, certifications, and real-world projects into a transparent evidence graph.
The end goal is not to create another score.
It is to help people build better work, prove what they can do, and see what they should build next.
Log in or sign up for Devpost to join the conversation.