AI & Data Science Specialization Course

This 18-month program is designed to transform BScCSIT and BCA students with primarily theoretical backgrounds into job-ready AI and Data Science professionals. The curriculum is structured around two phases: a 12-month shared foundation that builds core skills across all students, followed by a 6-month specialization phase where students choose one of two career tracks.
The program is deliberately practical-first. Every module is built around real datasets, real tools, and real business problems — with industry practitioners as module instructors and a full-time program coordinator ensuring continuity and quality across the entire 18 months.

Program structure at a glance

Weekly Schedule

Semester Calender

All 50 students follow the same curriculum during Phase 1. The goal is to build a strong, practical foundation across programming, data handling, statistics, machine learning, and cloud tools — the skills that are relevant regardless of which specialization track a student ultimately chooses.

SEMESTER 1 | Months 1–5 | Programming & Data Foundations

Semester 1 focuses on building the programming and data handling skills that every subsequent module depends on. Students with no prior Python experience will be brought to a professional working level. All work is done on cloud platforms from day one.

• Module 1 — Python programming foundations

(Duration: 5 weeks | Instructor: Industry practitioner (Python/backend developer)

• Module 2 — Data fundamentals with NumPy & Pandas

(Duration: 4 weeks | Instructor: Data engineer or senior data analyst)

• Module 3 — SQL & databases

(Duration: 4 weeks | Instructor: Database administrator or data engineer)

• Module 4 — Statistics for data science

(Duration: 4 weeks | Instructor: Data scientist with applied statistics background)

Semester 1 project

BREAK 1 | Month 6 | Exam break — structured async activities

No live sessions during break periods. Students are expected to complete structured async activities that build on semester 1 and prepare for semester 2. These are tracked and reviewed at the start of semester 2.

•  Complete Google Data Analytics Professional Certificate (Coursera) or IBM Data Science Professional Certificate — at minimum the first 3 courses
• Submit at least 1 Kaggle competition entry or notebook (any public competition)
• Polish semester 1 project based on coordinator feedback — update GitHub repository
• Read and summarize 2 industry articles on data trends in Nepal or global markets

SEMESTER 2 | Months 7–11 | Machine Learning & Analytics

Semester 2 introduces machine learning — the skill set that separates data-aware professionals from true practitioners. Every ML concept is introduced through a business problem first, then the algorithm, then the implementation. Students will also begin working with cloud deployment and basic MLOps.

• Module 5 — Data visualization & storytelling

(Duration: 3 weeks | Instructor: BI developer or data visualization specialist)

• Module 6 — Machine learning foundations

(Duration: 6 weeks | Instructor: Practicing ML engineer or senior data scientist)

• Module 7 — Cloud & MLOps fundamentals

(Duration: 3 weeks | Instructor: Cloud engineer or DevOps practitioner with ML experience)

• Module 8 — Business communication for data professionals

(Duration: 2 weeks | Instructor: Program coordinator or senior professional with stakeholder communication experience)

Semester 2 project

BREAK 2 | Month 12 | Track selection + certification

•  Complete AWS Cloud Practitioner (free prep available) or Microsoft Azure Fundamentals (AZ-900)
• Submit track selection form — Data Track or AI/ML Track — with a 200-word statement of intent
• Finalize GitHub portfolio: both S1 and S2 projects fully polished and documented
• LinkedIn profile setup and optimization — program coordinator reviews and provides feedback
• Attempt 2 Kaggle competitions or contribute to 1 open-source data project

At the start of month 13, students split into one of two tracks based on their interests, internship exposure, and performance in Phase 1. Both tracks run simultaneously with separate instructors. The program coordinator manages both tracks and ensures capstone quality across both groups.

SEMESTER 3A | DATA TRACK | Months 13–17

• Module D1 — Advanced SQL & analytics engineering

(Duration: 4 weeks | Instructor: Senior data engineer or analytics engineer)

• Module D2 — Data warehousing & cloud data platforms

(Duration: 3 weeks | Instructor: Data warehouse architect or cloud data platform specialist)

• Module D3 — ETL pipelines & data orchestration

(Duration: 3 weeks | Instructor: Data engineer with pipeline experience)

• Module D4 — Advanced BI & data visualization

(Duration: 3 weeks | Instructor: BI developer or Tableau/Power BI specialist)

• Module D5 — Data track capstone project

SEMESTER 3B | AI/ML TRACK | Months 13–17

• Module A1 — Deep learning foundations

(Duration: 4 weeks | Instructor: ML engineer or AI researcher with industry experience)

• Module A2 — NLP & large language model fundamentals

(Duration: 4 weeks | Instructor: NLP engineer or LLM practitioner)

• Module A3 — Computer vision basics

(Duration: 2 weeks | Instructor: Computer vision engineer)

• Module A4 — MLOps & production AI systems

(Duration: 3 weeks | Instructor: MLOps engineer or senior ML infrastructure practitioner)

• Module A5 — AI/ML track capstone project

BREAK 3 / MONTH 18 | Capstone polish + job readiness

Month 18 is the transition from student to job applicant. No new content is introduced. The focus is entirely on polish, preparation, and job search activation.

Job readiness activities

•  Capstone project: final polish based on industry partner and panel feedback
• GitHub portfolio: all 3 projects reviewed, READMEs finalized, pinned on profile
• LinkedIn profile: complete optimization — headline, About section, project showcases, skills
• Resume/CV: 1-page format tailored for both local and international applications
• Mock technical interviews: 2 rounds — one data/ML concepts, one coding/problem-solving
• Mock HR interviews: professional communication, salary negotiation basics
• Job application activation: apply to minimum 10 positions — local and remote
• Alumni network establishment: connect with industry partner contacts on LinkedIn

Assessment is designed to reflect how professionals are actually evaluated in the industry — not by their ability to recall theory under exam conditions, but by their ability to solve real problems and communicate results clearly.

• Overall grade composition

• Theory checkpoint policy

Theory checkpoints are gateway assessments — a student must score 60% or above before proceeding to the practical component of a module. They are not designed to be difficult. They validate that the student has understood the core concepts well enough to apply them. Students who fail may retake once after a 48-hour gap.

• Project evaluation rubric

The following tools represent the complete technology stack across the program. All tools have free tiers that cover everything students need. No paid software is required.

Program coordinator role

• The program coordinator (NextStep Infotech) is the full-time backbone of the program. This role is not a teaching role — it is a program management and quality assurance role.
•  Maintain curriculum continuity across all modules and instructors
• Track student progress and intervene early when students fall behind
•  Brief each incoming module instructor on student level and progress
• Manage industry partner relationships and source project datasets
• Run Sunday onsite lab sessions — workshops, mock interviews, guest talks
• Conduct monthly 1-on-1 check-ins with each student
• Manage theory checkpoint assessments and project evaluations

Module instructor guidelines

• Each module instructor is a working industry practitioner hired on a part-time basis for a specific module only. The following guidelines apply to all instructors:
• Instructors prepare their own module content within the framework provided by this syllabus
• All code examples must use real or realistic datasets — no toy datasets (no Titanic, no Iris for primary teaching)
•  Instructors must be available on the program Discord/Slack channel during their module for async questions
• Each instructor submits a module completion report to the coordinator at the end of their module
• Instructors are encouraged (not required) to attend at least one Sunday onsite session during their module

Guest session schedule

Each semester includes a minimum of 2 guest sessions during Sunday onsite time. These are separate from module instruction and are led by industry professionals sharing their career experience.

• Industry partner tier model

• Capstone project guidelines

• Both track capstones must meet the following minimum standards to qualify as industry-grade deliverables:
• Problem statement must be provided or approved by an industry partner — not invented by students or coordinator
• Dataset must be real or realistic — synthetic data is permitted only if it accurately mirrors real-world patterns
• All code must be in a public GitHub repository with full documentation
• Final presentation must be attended by at least one industry partner representative
• Students must be able to explain every technical decision made in their project
• Projects are presented publicly — students should be proud to share the link on LinkedIn

College admission prerequisites

The following requirements must be communicated to students and enforced by the college administration before program enrollment:

Student expectations

• Attend minimum 80% of live sessions — below this threshold may result in removal from the program
• Submit all project deliverables on time — late submissions lose 10% per day, not accepted after 3 days
• Maintain an active GitHub profile — all work must be committed, not stored local
• Participate in Sunday onsite sessions — minimum 75% attendance required
• Complete async certifications during break periods — proof of completion required
• Engage respectfully with instructors, peers, and industry partners at all times

Upon successful completion of this program, graduates will be able to:

• Write production-quality Python code for data manipulation, analysis, and ML model development
•  Design and query relational databases and cloud data warehouses using SQL
• Build, evaluate, and deploy machine learning models as REST APIs
• Work comfortably on cloud platforms including AWS and GCP
• Communicate data findings clearly to both technical and non-technical audiences
• Maintain a professional GitHub portfolio demonstrating real project experience

Additionally, Data Track graduates will be able to:

• Build and maintain production ETL pipelines using Airflow and dbt
• Design data warehouse schemas and implement them on cloud platforms
• Build business intelligence dashboards in Power BI and Tableau

Additionally, AI/ML Track graduates will be able to:

• Train and fine-tune deep learning models using PyTorch
• Build NLP pipelines and integrate large language model APIs
• Deploy AI applications with monitoring and responsible AI documentation