Learning routes / Applied Generative AI and AI Development
Learning routeApplied Generative AI and AI Development
Build and evaluate practical AI applications. Learners who can already program in Python and want to build AI applications.
- Length
- 16 weeks · 160 training hours
- Weekly effort
- About 8–12 hours a week
- Start with
- Working Python: functions, files, JSON and calling a web API. An entry check is used before Generative AI.
- Language
- Taught in English. Practice questions and explanations are in English.
What you will be able to do
Builds a document assistant with model APIs, retrieval and source attribution, measures its quality, and documents its failure cases and limits.
Be clear: Builds practical AI application skills. It is not a machine-learning research or AI-engineer qualification.
The route, step by step
- Optional
Python — Beginner
Only if you do not yet meet the Python entry check.
Portfolio Project: Shop sales tracker
How it is checked: Portfolio project output: Working tested software application with source code and technical README. Completion is subject to the course-specific proposed policy below; practical evidence is mandatory. Rubric focus: Code quality, security, testing and demonstration. Institute certificate only; external/vendor certification separate. Independent final check: The supplied file and script run verbatim: the valid pen row contributes 30, the invalid quantity is reported, and missing headers are diagnosed. Proposed completion: 70% overall and 70% in the portfolio practical; critical accuracy, permission and privacy failures require correction and observed reassessment. This policy needs academic approval before use for awards.
- Start here
Generative AI
Model APIs, structured outputs, tool use, retrieval and evaluation in one guided project.
Portfolio Project: Evaluated AI document assistant with approval workflow
How it is checked: Portfolio project output: Evaluated AI document assistant with approval workflow. Suggested passing criteria: 70% suggested (60% for entry foundation / short interventions); practical evidence mandatory. Rubric focus: Grounded-answer evaluation + security test. Institute certificate only; external/vendor certification separate.
- Then
RAG Quality and Evaluation
Measuring retrieval and answer quality with evidence, not impressions.
Portfolio Project: Evaluated document assistant with retrieval comparison
How it is checked: For academic approval: 70% overall and 70% Portfolio Project. Correct critical evidence, privacy or permission failures and complete an observed reassessment. Shared responsible-use questions supplement, rather than replace, course-specific practical assessment.
- Optional
AI Agents and Automation Builder
Bounded agents and workflows with approval gates, for those who want automation.
Portfolio Project: Customer support agent with approved knowledge and refund approval
How it is checked: Practical assessment uses the learner's own agent: tool tests, an evaluation harness run, prompt-injection test results, an approval demonstration and a sandbox deployment.
- Optional
AI Application Security and Reliability
Prompt injection, unsafe output handling and reliability checks for AI applications.
Portfolio Project: Authorised AI security and reliability test harness
How it is checked: For academic approval: 70% overall and 70% Portfolio Project. Correct critical evidence, privacy or permission failures and complete an observed reassessment. Shared responsible-use questions supplement, rather than replace, course-specific practical assessment.
Certificate
ADIVAS issues an institute certificate for each programme when its requirements are met: the course weeks, the graded quizzes and, where recorded, attendance. It is an ADIVAS certificate, not an external or vendor qualification, and anyone can check it with its code.
Career support
Placement assistance for college and working learners: resume builder, AI mock-interview practice, Skill Passport and openings shared with ADIVAS.
We prepare you for employment; we do not guarantee employment.
Batches
No batch for this route is scheduled on the website yet. A counsellor can tell you when the next one starts.
Questions
- How long does this route take?
- The required steps add up to 16 weeks and 160 training hours (about 8–12 hours a week). Optional steps are extra.
- What do I need before I start?
- Working Python: functions, files, JSON and calling a web API. An entry check is used before Generative AI.
- Will this get me a job?
- Builds practical AI application skills. It is not a machine-learning research or AI-engineer qualification. Placement assistance for college and working learners: resume builder, AI mock-interview practice, Skill Passport and openings shared with ADIVAS.
- How much does it cost?
- Fees depend on the programmes and batch you choose. A counsellor shares current fees, batch timings and a suitable learning plan when you ask.
Curriculum outlines are drafts under academic review; a counsellor confirms what the current batch covers. Other routes: AI and Digital Explorer for Children, School AI and Coding Foundations, Software Development for College and Career Learners, Data Analytics and Business Intelligence.