Inspiration
🌟 What Inspired Us
The idea for FlowFix began with a real incident in my own family. My cousin, who was looking for internships on LinkedIn, received an email offering her a “guaranteed internship”—but only if she paid a security deposit of ₹25,000.
This was a shocking moment. She was excited, the offer looked professional, and the email had a company logo. But something felt off.
After checking the company domain, reviewing the sender email, and doing a bit of online research, we realized:
It was a scam targeting young students.
We reported it, but the experience made one thing clear: Students desperately need a simple tool to verify internship authenticity. That incident became the spark that led to FlowFix.
What it does
FlowFix is a simple, student-friendly tool that helps users instantly check whether an internship posting is real or suspicious. Students paste any internship link—from LinkedIn, Instagram, Telegram, or job portals—and FlowFix automatically scans it for known scam signals.
The tool analyzes the link using three layers:
Domain Safety Check – verifies the website’s trustworthiness.
Red-Flag Keyword Detection – catches phrases like “security deposit,” “processing fee,” or “guaranteed selection.”
Trust Score (0–100) – gives a clear risk rating so students can decide confidently.
FlowFix also includes community reporting and a verified internship directory, helping students avoid fraud and discover safe opportunities. Overall, it reduces time wasted on fake offers and protects students from financial and identity risks.
How we built it
🛠 How I Built the Project
- Research Phase
Collected 200+ real internship links (genuine + scam).
Identified keywords, patterns, and trigger phrases used in scams.
- Architecture & Features
I planned the system around three pillars:
URL Scanner — checks domain legitimacy
Keyword Engine — detects payment requests, unrealistic claims, etc.
Trust Score Generator — produces a simple 0–100 score
- User Interface
Designed a mobile-first layout using simple HTML/CSS mockups.
Users can paste a link and instantly see results.
- Testing
Collected feedback from students in my school.
Adjusted scoring based on common false positives.
Challenges we ran into
⚠️ Challenges I Faced
- Distinguishing Scams from Poorly Written Genuine Posts
Some real internships look unprofessional. Some scams look extremely polished. This made detection tricky.
- Getting Reliable Data
There are many scam stories online but very few structured datasets. I had to manually collect examples from:
Instagram pages
Telegram job channels
News articles
- Designing a Fair System
I had to ensure the tool doesn’t falsely accuse legitimate companies. So instead of saying “Scam”, FlowFix uses probability scoring, such as:
Risk Level
78 % Risk Level=78%
This avoids mislabeling and keeps results ethical.
Accomplishments that we're proud of
Accomplishments I Am Proud Of
Turning a real-life problem into a real solution: FlowFix was inspired by an actual scam attempt involving my cousin. Transforming that experience into a tool that can protect hundreds of students is something I’m genuinely proud of.
Building a functioning trust-score model: Creating a scoring formula that evaluates domains, keywords, and scam indicators helped me understand real-world safety systems and apply basic mathematical weighting.
Collecting and analyzing real scam data: Instead of relying on examples from AI, I manually gathered genuine internship links, scam posts, and news cases to make the tool realistic and evidence-based.
Designing a clean, student-friendly interface: The mobile-first layout ensures that any student can quickly check an internship link with no learning curve.
Prioritizing ethics and fairness: Instead of labeling anything as “scam,” FlowFix uses probabilistic risk levels, preventing false accusations and maintaining responsible use.
Staying aligned with the theme “Friction Points”: FlowFix directly tackles a major real-world friction—digital trust and scam prevention—especially for first-time internship seekers.
What we learned
📚 What I Learned
Working on FlowFix taught me lessons in:
- Understanding Real-World Scams
How scam recruiters use identical tactics: urgency, advance fees, fake domains.
How easy it is to impersonate a company online.
- Research & Data Verification
I learned to look for credible datasets, surveys, and reports.
I used real statistics like the 379% rise in internship scams in India (2023).
- Designing User-Focused Solutions
The tool must be simple enough for a 15-year-old first-time applicant.
I studied friction points in platforms like Instagram, Telegram, and WhatsApp job groups.
- Breaking Down Problems Mathematically
While building the trust score model, I began experimenting with weight-based scoring:
Trust Score=100−(w1k1+w2k2+w3k3+…)
where 𝑘𝑖 are red-flag indicators and 𝑤𝑖 are their assigned risk weights. This helped me understand how scoring models work in real safety tools.
What's next for FlowFix
What's Next for FlowFix
The current version of FlowFix is a strong starting point, but there are several upgrades planned to make it more accurate, reliable, and widely accessible:
- AI-Based Scam Detection
Integrate a lightweight ML model trained on verified scam datasets to move from rule-based checks to smarter, context-aware analysis.
- Browser Extension
Build a Chrome/Edge extension that automatically flags suspicious internship pages on LinkedIn, Instagram, and job portals—no need to copy-paste links.
- Community-Verified Internship List
Allow students to upvote/downvote internship postings and companies, creating a constantly updated trust ecosystem.
- Scam Pattern Database
Create a public, searchable library of common scam messages, email templates, Telegram posts, and fraud domains to spread awareness.
- Collaboration with Schools & NGOs
Partner with student communities, cyber safety clubs, and youth organizations to promote safer internship searching.
- In-App Reporting to Cyber Cells
Enable one-click reporting to authorized cyber safety channels to support real-world action against scammers.
Log in or sign up for Devpost to join the conversation.