Demo Mode

No student ID available

Certification - AI-2.5 Modern Artificial Intelligence | Telebort Coding Education

Certification Requirements

Certification Requirements: AI-2.5 Modern Artificial Intelligence

Overview

To earn your AI-2.5: Modern Artificial Intelligence certification, you must demonstrate mastery of reinforcement learning and generative AI through coursework, projects, and assessments.

Completion Criteria

One. Lesson Completion (30%)

Requirements:

Complete all 18 lessons (Concepts 01-18)
Submit lesson reflections (1-2 paragraphs per lesson, submitted weekly)
Participate in discussion forums (minimum 10 substantive posts throughout course)
Attend or watch recordings of weekly office hours (recommended, not required)

Grading:

Lesson completion: 15% (binary: complete or incomplete)
Reflections: 10% (quality and depth of understanding)
Discussion participation: 5% (helpfulness and engagement)

Reflection Prompts (due with each lesson):

What was the most challenging concept in this lesson?
How does this lesson connect to previous lessons or real-world applications?
What question do you still have after completing the lesson?

2. Activities (25%)

Requirements:

Complete all 18 hands-on activities (Activities 01-18)
Achieve 70% or higher on success criteria for each activity
Submit within 1 week of lesson completion
Include brief implementation notes explaining your approach

Grading:

Code completion: 40% (all TODOs implemented correctly)
Code quality: 30% (clean, readable, well-commented)
Visualizations: 20% (plots display correctly with proper labels)
Understanding: 10% (brief notes demonstrate comprehension)

Submission Format:

Jupyter notebook (.ipynb) with all cells executed
Output visible for all cells
Brief markdown cell at end summarizing approach and results

Late Policy:

1-3 days late: -10%
4-7 days late: -25%
7 days late: Requires instructor approval

3. Projects (30%)

Requirements:

Complete at least 6 of 7 projects (Project 7 Capstone is required)
Meet all technical requirements per project rubric
Submit before project deadline (typically 2-3 weeks per project)
Achieve 75% average across all submitted projects

Project Deadlines:

Project	Release	Due	Duration
One: DQN Game Master	After Lesson 3	2 weeks	2 weeks
2: Autonomous Robot Navigation	After Lesson 6	2 weeks	2 weeks
3: GAN Art Studio	After Lesson 12	2 weeks	2 weeks
4: Latent Space Explorer	After Lesson 10	2 weeks	2 weeks
5: Text Generation with RLHF	After Lesson 16	3 weeks	3 weeks
6: Multi-Modal Content Generator	After Lesson 17	3 weeks	3 weeks
7: Capstone - AI Agent Ecosystem	After Lesson 18	4 weeks	4 weeks

Grading Rubric (per project):

Functionality: 40% (works as specified, meets requirements)
Code Quality: 25% (clean, modular, documented)
Technical Depth: 20% (demonstrates understanding of algorithms)
Creativity: 10% (originality in approach or extensions)
Documentation: 5% (README, comments, submission report)

Project Submission Requirements:

Complete Jupyter notebook or code repository
README.md with setup instructions
Portfolio output (agent video, generated samples, demo, etc.)
Brief technical report (500-1000 words) explaining:
- Implementation approach
- Challenges encountered and solutions
- Results and performance analysis
- Potential improvements

4. Assessments (15%)

Requirements:

Pass all 18 Exit Tickets (70% average across all lessons)
Pass Mid-Course Assessment (70% minimum, covers Lessons 1-9)
Pass Final Assessment (75% minimum, covers entire course)

Exit Tickets:

5 questions per lesson (90 questions total)
Auto-graded online quizzes
Unlimited attempts, best score counts
Must complete before moving to next lesson

Mid-Course Assessment (After Lesson 9):

Format: Jupyter notebook + multiple choice
Duration: 90 minutes
Coverage: Reinforcement Learning module (Lessons 1-8)
Weight: 5% of total grade
Content:
- 15 multiple choice (conceptual RL questions)
- Code completion: Implement Q-Learning agent
- Debugging: Fix broken DQN implementation
- Analysis: Compare RL algorithms on given task

Final Assessment (After Lesson 18):

Format: Jupyter notebook + essay
Duration: 120 minutes
Coverage: Entire course, emphasis on integration (RLHF, multi-modal)
Weight: 10% of total grade
Content:
- 20 multiple choice (GenAI and RLHF concepts)
- Implementation: Build simplified RLHF pipeline
- Debugging: Fix GAN training instability
- Essay: "How RL and GenAI converge in modern AI" (500-750 words)

Retake Policy:

Exit Tickets: Unlimited attempts
Mid-Course Assessment: 2 attempts allowed
Final Assessment: 2 attempts allowed
Must wait 48 hours between attempts

Grading Breakdown

Component	Weight	Minimum Required
Lessons & Reflections	30%	70% average
Activities	25%	70% each activity
Projects	30%	75% average (6 of 7)
Assessments	15%	70% (Exit Tickets, Mid-Course), 75% (Final)
Total	100%	75% overall

Final Grade Calculation:

scss

Final Grade = (Lessons × 0.30) + (Activities × 0.25) + (Projects × 0.30) + (Assessments × 0.15)

Grade Thresholds:

A: 90-100%
B: 80-89%
C: 75-79% (minimum for certification)
F: Below 75% (no certification)

Time Commitment

Recommended Schedule

Lessons: 3-4 hours per lesson (theory + activity)
Projects: 8-12 hours per project
Assessments: 2-4 hours total (Exit Tickets + exams)

Total Estimated Time: 120-150 hours

Pacing Options

Standard Pace (16 weeks):

2 lessons per week
Start projects as soon as unlocked
Complete 1-2 projects every 3-4 weeks
8-10 hours per week

Accelerated Pace (10 weeks):

3 lessons per week
Parallel project work
12-15 hours per week
Recommended for students with strong RL/ML background

Relaxed Pace (20 weeks):

1 lesson per week
More time for projects and mastery
6-8 hours per week
Good for working professionals

Prerequisites Check

Before enrolling, ensure you meet these prerequisites:

Required

Completed AI-2 (or equivalent ML/CV course)
Solid Python programming skills
Understanding of:
- Neural networks and backpropagation
- Gradient descent optimization
- NumPy, Matplotlib basics
Linear algebra (vectors, matrices, dot products)
Basic calculus (derivatives, chain rule)

Certificate Benefits

Upon successful completion, you will receive:

Digital Certificate

LinkedIn-compatible credential
Verifiable via Telebort Engineering portal
Shows completion date and final grade

Portfolio

6-7 production-quality AI projects:
- Game-playing agents (DQN)
- Autonomous navigation (PPO)
- Generative art (GANs)
- Image manipulation (VAEs)
- LLM alignment (RLHF)
- Multi-modal systems (CLIP + Diffusion)
- Capstone AI agent ecosystem
GitHub-ready code repositories
Shareable demos and videos

Career Advancement

AI Specialist Badge on Telebort platform
Access to AI-3 (Applied Generative AI) course
Letter of Recommendation (upon request, 90%+ grade)
Alumni Network access for career opportunities

Skills Mastered

Reinforcement learning (Q-Learning, DQN, Policy Gradients, PPO)
Generative models (VAEs, GANs, Diffusion, Transformers)
RLHF and AI alignment techniques
PyTorch and modern AI frameworks
Production AI development and deployment

Support Resources

Office Hours

Schedule: Tuesdays and Thursdays, 6-7 PM ET
Format: Zoom (live Q&A, code walkthroughs)
Recording: Available for 48 hours

Discussion Forums

Platform: Discord server (invite sent upon enrollment)
Channels:
- #lessons-rl (Lessons 1-8 discussion)
- #lessons-genai (Lessons 9-15 discussion)
- #lessons-rlhf (Lessons 16-18 discussion)
- #projects (project help and showcase)
- #technical-help (debugging and environment setup)
- #general (course logistics and announcements)