SkillBridge - AI Career Operating System

The AI-powered career OS for tech students. Daily missions, skill gap analysis, AI coaching, job matching & interview prep ; all connected. Your path from student to hired, personalized.

Comment

💡 Inspiration

"73% of tech students apply to 100+ jobs and never hear back. Not because they aren't smart enough. Because nobody told them exactly what to fix."

I spent months preparing for Data Science and Machine Learning interviews completely alone.

No mentor. No personalized roadmap. No feedback on whether I was actually ready.

I fixed my resume more than 17 times. I applied to 100+ positions with a response rate under 3%. I studied 12+ hours per week with no idea if I was preparing the right things. I practiced SQL with nobody to evaluate my answers. I read research papers at midnight wondering if I was even preparing the right things.

The tools I found were either too generic, too expensive, too passive, or completely disconnected from each other:

Tool What It Gave Me What It Could Not Give Me
Coursera Course completion % Whether I was interview-ready
LinkedIn Job listings My actual match score
LeetCode Problem rankings Career context for my role
YouTube Content Personalization
Notion Organization Insight

None of them talked to each other. None of them knew my situation. None of them told me what to do next.

I built SkillBridge at 2 AM because I was tired of preparing alone. And I refused to believe I was the only one.

This is not just a hackathon project. It is the product I desperately needed and could not find anywhere at any price. The problem was never effort. The problem was direction.

$$ \text{Interview Success} = f\Bigl(\text{Right Skills} \times \text{Preparation Quality} \times \text{Application Strategy}\Bigr) $$

Most students optimize only one variable. SkillBridge optimizes all three simultaneously.

What SkillBridge Does

SkillBridge is a complete AI-powered career operating system where six career tools are deeply connected into one ecosystem.

$$ \text{SkillBridge} = \text{Notion} + \text{LinkedIn} + \text{Coursera} + \text{Duolingo} + \text{LeetCode} + \text{AI Mentor} $$

Built specifically for students targeting high-growth tech careers in 2026:

Target Role Key Skills Avg Fresher Salary India 2026
Data Scientist Python, SQL, Statistics, ML, Deep Learning 8-15 LPA
ML Engineer PyTorch, MLOps, Docker, Cloud, Deployment 8-16 LPA
AI Engineer LLMs, RAG, LangChain, NLP, Vector DBs 10-18 LPA
Python Developer FastAPI, DSA, System Design, Testing 6-12 LPA
Data Analyst SQL, Power BI, Tableau, Statistics, EDA 4-8 LPA

India alone needs 1.2 million AI professionals by 2027. Current supply is only 420,000. SkillBridge exists to close that gap one student at a time.

Although the demo video primarily showcases the Python Developer pathway, SkillBridge is designed as a multi-role AI career operating system supporting a wide range of high-growth technology careers including Data Scientist, ML Engineer, AI Engineer, Data Analyst, Python Developer, and future AI/ML specializations. The underlying architecture, Skill Gap Map, Job Match system, Interview War Room, and Aria AI Coach dynamically adapt based on the selected career path, allowing the platform to personalize guidance for different domains across AI, Machine Learning, Data Science, Backend Development, NLP, and emerging Generative AI roles.

Six Fully Connected Systems

📊 1. Personalized Dashboard

The dashboard acts as a real-time career readiness monitor.

It tracks:

  • Career Match Score
  • XP and Level
  • Daily missions
  • Study streak
  • Readiness growth
  • Skill verification progress

The readiness score is mathematically calculated:

$$ R = \frac{\sum_{i=1}^{n} w_i \cdot p_i}{\sum_{i=1}^{n} w_i} \times 100\% $$

Where:

  • (w_i) = criticality weight of skill (i)
  • (p_i) = progress on skill (i)

Weights:

$$ w_i = \begin{cases} 0.75 & \text{CRITICAL skill} \ 0.20 & \text{MODERATE skill} \ 0.05 & \text{NICE TO HAVE} \end{cases} $$

2. Skill Gap Map

The Skill Gap Map shows exactly:

  • Which skills users already have
  • Which skills are blocking interviews
  • Which skills should be prioritized next

Each skill card includes:

  • Progress slider
  • Status tracker
  • Days-to-close estimate
  • Verified resources
  • XP rewards

Skills are universal across roles. SQL progress remains synced even if users switch career paths.

3. Study Planner

This is the feature that completely changed the project.

Users can:

  • Start live study sessions
  • Track exact study time
  • Update progress instantly
  • Earn XP
  • Maintain streaks
  • Sync progress across the whole platform

Study sessions trigger a complete platform-wide cascade:

User studies SQL for 45 minutes
             ↓
SQL progress increases
             ↓
Career readiness updates
             ↓
XP and streak increase
             ↓
Aria changes recommendations
             ↓
Job Match updates verified skills

XP system:

$$ \text{XP}_{\text{session}} = \begin{cases} 25 & t < 30\text{ min} \ 50 & 30 \leq t < 60\text{ min} \ 100 & 60 \leq t < 120\text{ min} \ 200 & t \geq 120\text{ min} \end{cases} $$

4. Job Match Analyzer

Users paste any job description and SkillBridge calculates:

  • Match percentage
  • Missing skills
  • Recommended actions
  • Interview readiness

$$ M = \frac{|S_{\text{user}} \cap S_{\text{required}}|}{|S_{\text{required}}|} \times 100\% $$

Recommendations:

  • Apply Now
  • Apply with Gap Prep
  • Build Skills First

5. Interview War Room

The Interview War Room contains:

  • 200+ interview questions
  • DSA rotation system
  • ML theory challenges
  • System design prompts
  • FAANG-level mock preparation

Adaptive difficulty:

$$ q_{\text{next}} = \arg\max_{q \in Q_{\text{harder}}} \left[ \text{Relevance}(q,\text{role}) - \text{Seen}(q,\text{user}) \right] $$

Questions never repeat once completed.

6. Aria AI Career Coach

Aria is not a generic chatbot.

She dynamically adapts using:

  • Current readiness score
  • Skill gaps
  • Last study session
  • XP level
  • Target role
  • Dream company
  • Streak data

Example:

"Emergency mode activated. For your Google Data Scientist interview tomorrow, focus first on SQL window functions and ML evaluation metrics. Then practice your strongest project explanation in under 3 minutes."

How We Built It With MeDo

The Core Innovation

SkillBridge was built entirely through MeDo using natural language conversations.

No traditional IDE.
No manual frontend setup.
No backend boilerplate.
No writing thousands of lines of code manually.

Everything was iterated conversationally through MeDo across 62 iterations in 25 days.

This hackathon project specifically explores how far natural language application development can go when combined with strong system design thinking.

How MeDo Was Used

Phase Iterations What MeDo Generated
Foundation v1-v5 Dashboard, onboarding, database schema
Core Systems v6-v15 Skill Gap Map, Job Match, War Room
Intelligence v16-v22 Aria AI personalization
Cross-System Sync v23-v28 Study Planner live updates
Polish v29-v62 Animations, auth, UI refinement, celebrations, polishing

Most Impressive Thing MeDo Generated

This was the single most impressive MeDo generation during development:

"When a study session ends, update that skill's progress in the database, recalculate career readiness, increase XP, increment streaks, sync the update across Dashboard and Skill Gaps, and make Aria automatically recommend the next biggest weakness."

MeDo correctly generated:

  • Cross-page state updates
  • Database synchronization
  • Dynamic recommendation logic
  • XP calculations
  • Persistent session tracking
  • UI updates across multiple systems

That single prompt connected six independent systems together.

Traditional implementation time would likely take multiple days for an experienced engineer.

MeDo Features Used

  • [x] Multi-turn conversational app development
  • [x] Full-stack generation
  • [x] Frontend UI generation
  • [x] Backend logic generation
  • [x] Database schema generation
  • [x] Authentication setup
  • [x] Supabase integration
  • [x] Cross-page state management
  • [x] Animation generation
  • [x] Real-time synchronization logic
  • [x] Visual refinement tools

Tech Stack Generated Through MeDo

Layer Technology
AI Builder Platform MeDo
Frontend React + TypeScript
Styling Tailwind CSS
Animations Framer Motion
Database Supabase PostgreSQL
Authentication Supabase Auth
Deployment MeDo Deploy

Challenges We Faced

Aria Memory Synchronization

Early versions of Aria responded with repetitive generic answers because state updates were cached incorrectly.

The fix:

// Wrong
const response = generateResponse(message, cachedProfile)

// Correct
const profile = await fetchUserProfile(userId)
const response = generateResponse(message, profile)

Cross-System Synchronization

The hardest engineering challenge was ensuring:

  • Study Planner
  • Dashboard
  • Skill Gap Map
  • Job Match
  • Aria
  • XP System

all updated from one action without inconsistencies.

This required careful database architecture and precise prompting inside MeDo.

Building Under Credit Constraints

Every deep MeDo generation consumed credits.

That forced highly optimized prompting and architectural planning rather than random experimentation.

Ironically, this constraint made the final product cleaner and more coherent.

Accomplishments We Are Proud Of

  • Built a production-level AI career platform in 25 days
  • Created six deeply connected systems
  • Achieved real-time cross-platform synchronization
  • Built entirely through conversational development
  • Designed a psychologically motivating learning loop
  • Developed a personalized AI mentor system

The moment I knew it worked:

I completed a SQL study session, watched my readiness increase live, and Aria immediately changed her recommendations based on my updated profile.

That felt real.

📚 What We Learned

MeDo fundamentally changed how we think about software development.

The hardest skill was not coding.

It was:

  • clarity of vision
  • system thinking
  • precision in prompting
  • iteration discipline

$$ \text{Time to MVP} = f\left( \frac{\text{Clarity of Vision}} {\text{Technical Barrier}} \right) $$

With tools like MeDo, the technical barrier is shrinking rapidly.

The remaining bottleneck is imagination and clarity.

🔮 What's Next

  • 📱 Mobile application
  • 👥 Peer accountability groups
  • 🏅 Verified skill badges
  • 🔗 LinkedIn integration
  • 🎯 Mentor matching
  • 🏫 University dashboards
  • 🌍 Multi-language support
  • 📄 AI-powered resume builder

❤️ Final Note

SkillBridge exists because preparing for tech careers should not feel lonely.

Every student deserves:

  • direction
  • mentorship
  • clarity
  • feedback
  • confidence

And that is exactly what SkillBridge is trying to build.

Built with MeDo.

Built With

  • ai-powered-career-recommendation-system
  • framer-motion
  • medo-(baidu-ai-application-builder)
  • prompt-engineering
  • react
  • real-time-state-synchronization
  • supabase-authentication
  • supabase-postgresql
  • tailwind-css
  • typescript
Share this project:

Updates