A hands-on agentic AI program
Make AI do real work.
60 hours30+ projectsBuilt by you
30+ reasons
Bring a task you understand. Explore how connected tools and useful AI can turn it into a build you can test and explain.
At least 30 projects across the program. The mix evolves with each group's backgrounds, experience and learning needs.
Project scope varies. Not every build is a standalone deployed application.
60hours of
making.
Learn. Build. Review. Repeat.
Connect automation, retrieval, agents and deployment in one practical journey. Each session adds working capability you can understand, test and demonstrate.
Arabic explanation. English materials.
Your destination / In everyday terms
Start with a familiar business task. Learn to build a focused solution, check its results and explain how it works.
Build a flow that captures an enquiry, creates a record and prepares the next follow-up.
Build an assistant that retrieves relevant information and shows the sources behind its answer.
Extract fields or summarize a document, then check the result before someone relies on it.
Create a simple interface or shareable result and demonstrate a complete, tested user journey.
The program
12 tools & concepts. One connected learning journey.
Let one app send information to another, without copying it by hand.
A new enquiry arrives in one place and becomes a structured record in another.
Use visual steps to move a task from an incoming event to a useful result.
Connect services using a visual workflow editor.
A follow-up flow that captures an enquiry, routes it and prepares the next action.
Build flexible workflows with self-hosting options.
A document intake flow that extracts information, checks it and routes exceptions.
Find relevant information by meaning, not only exact words.
A searchable knowledge collection that finds passages related to a question.
Give the model relevant source material before it answers.
A document assistant that answers with sources and flags questions that need review.
Run a suitable AI model on your own machine or server.
A local AI step that summarizes or extracts information from a sample document.
Turn prompts and knowledge into an assistant people can use.
A usable assistant interface with instructions, knowledge and a testable workflow.
Work with an AI coding assistant.
A small feature or workflow helper you can test and explain.
Give an agent controlled access to useful tools.
An assistant that retrieves information or calls an approved action.
Supervise a multi-step task from an objective.
A supervised task that produces a reviewable research or work summary.
Build and refine an application interface from a clear brief.
A front end where a colleague can submit a request and see the result.
Publish a suitable application and manage its configuration.
A working application link you can share, test and demonstrate.
The session experience
Try it.Build it.Break it.Improve it.
Every session includes a practical project.
See the concept. Build with it. Inspect the result. Debug what fails and carry the learning into the next layer.
60 hours. Time follows the group's learning needs.
Roadmaps / Start without a technical background
One common core, different starting points. These milestones describe the learning journey, not separate programs or fixed weekly schedules.
Choose a repetitive job, identify its inputs and define what a successful result looks like.
Move information between apps and build a rule-based workflow.
Use documents, summaries or extraction where language understanding helps.
Test examples, inspect failures and decide when a person must approve the next action.
Add a simple interface or shareable output and demonstrate the complete flow.
Roadmaps / Bring your starting point
Start with
A process you understand
Demonstrate a repeatable process with clear exception handling.
Start with
A problem worth prototyping
Show an early solution and explain what further development needs.
Start with
A service you can demonstrate
Present a portfolio example with a clear scope and business benefit.
Start with
An existing system to extend
Demonstrate a scoped agent integration and inspect its behavior.
Your group helps shape the examples and project scope. No previous advanced programming experience is required; guided technical practice is part of the journey.
Built by you. Shaped by your group.
Illustrative build directions. Projects vary by group. Connect, retrieve, automate and build. The exact projects reflect participants' backgrounds and learning needs.
Projects / Example build menu
30+ projects across the program, with a practical project in every session. The 15 examples below are a selection from the possible builds; the final mix varies by group.
Turn a submitted form into a clean record ready for follow-up.
Send each request to the appropriate team using clear rules.
Create the next reminder or draft response from an incoming enquiry.
Turn notes into a structured summary and an action list for review.
Pull selected fields from a sample document into structured output.
Find relevant passages from a small approved knowledge collection.
Answer questions using selected documents and show the sources.
Test a document assistant and record where answers need improvement.
Each example is defined by a visible result. Projects vary in size; these are not promises of 30 full applications or coverage of every industry.
Run a suitable model on available hardware and inspect its summary.
Create a simple place for a user to ask and receive grounded answers.
Let an agent retrieve information through a scoped tool connection.
Pause before an important action so a person can review the result.
Produce a source-backed work summary and check it before use.
Give a colleague a simple interface to submit a request to a workflow.
Deploy a suitable build and demonstrate the full user journey.
Market applications
Examples of where agentic AI can help at work. These are market use cases, not a list of course modules or guaranteed projects.
Find a policy clause and draft a reply with its source.
Check a submitted file for missing documents before staff review.
Extract invoice details and flag mismatches for the accountant.
Prepare appointment reminders and organize booking requests.
Answer product questions and retrieve order updates.
Match an enquiry to suitable listings and prepare a follow-up.
Turn a new lead into a CRM record and a follow-up draft.
The examples illustrate possibilities. Real use depends on suitable data, permissions and human review.
Turn approved lesson notes into practice questions for teacher review.
Turn a campaign brief into draft posts for the team to review.
Answer onboarding questions using approved company policies.
Summarize shipment updates and flag delayed deliveries.
Find a documented answer and pass complex cases to the right person.
Find relevant contract clauses and prepare a summary for lawyer review.
Compare supplier quotes and highlight price or delivery differences.
The examples illustrate possibilities. Real use depends on suitable data, permissions and human review.
Turn fault reports into a clear maintenance task list.
Draft guest replies using booking details and approved policies.
Group customer feedback and prepare a daily issues summary.
Classify support tickets and suggest steps from the help library.
Turn site notes into a progress report with owners and next steps.
Organize volunteer applications and draft scheduling messages.
One skillset. Different business contexts. Start with a repeated task, connect the right information, then check the result. Course projects are adapted to each group's background.
Illustrative applications
Customer-facing teams
Finance & purchasing
Support & operations
Customer-facing teams · Less searching. Clearer follow-up.
The useful question is simple: where does your team keep copying information, searching for answers or preparing the same response?
A website form with a name, contact details and a product question.
Capture the enquiry, check required fields, create a CRM record and prepare a reply draft. Route incomplete requests for review.
One organised lead record with an owner and a suggested next step.
A salesperson checks the details and approves the message before sending.
Less retyping between the enquiry form and the sales system.
APIs & webhooksMake or n8n
Customer-facing teams · Less searching. Clearer follow-up.
The useful question is simple: where does your team keep copying information, searching for answers or preparing the same response?
An employee question and an approved set of company policies or product guides.
Find relevant document passages and draft an answer that links back to them. Flag missing or conflicting information.
A useful answer alongside the source passages used to prepare it.
The reader checks the source; a responsible person handles unclear cases.
A clearer route to information already held inside the business.
Vector databasesRAGDify
Finance & purchasing · Paperwork with a clearer next step.
AI can help prepare information for a decision. The person responsible still checks the evidence and makes the decision.
Sample supplier invoices and a list of fields the accounting team needs.
Extract supplier, invoice number, date and totals into structured fields. Flag missing values and inconsistent totals.
A review table that keeps each extracted record linked to its source.
An accountant checks the figures before posting or approving payment.
Less manual copying and a visible list of records that need attention.
n8nDocument extractionOllama, where hardware fits
Finance & purchasing · Paperwork with a clearer next step.
AI can help prepare information for a decision. The person responsible still checks the evidence and makes the decision.
Quotes for the same purchase, with quantity, price and delivery terms.
Extract comparable fields, keep units and currencies visible and highlight missing terms. Prepare a comparison for the buyer.
A side-by-side summary with references to the original quotations.
The buyer verifies the terms and chooses the supplier.
A consistent first review before the commercial discussion.
Document extractionMake or n8nRAG for source lookup
Support & operations · A request arrives. The right work follows.
Connected workflows make handovers easier to follow. Useful interfaces help colleagues provide the right information at the start.
An incoming IT support request and an approved troubleshooting library.
Classify the request, find relevant guidance and draft a suggested response. Route uncertain cases to the right support queue.
A ticket summary, relevant help sources and a response ready to review.
A support specialist approves the response and any system action.
Less time gathering context before a specialist takes over.
n8nRAGMCP for scoped tool access
Support & operations · A request arrives. The right work follows.
Connected workflows make handovers easier to follow. Useful interfaces help colleagues provide the right information at the start.
A colleague submits a structured request through a simple web form.
Validate the fields, trigger the connected workflow and prepare a result for review. Show a clear confirmation to the requester.
A request record and a visible result the team can inspect.
A named owner reviews exceptions and approves important actions.
A repeatable handover with fewer missing details.
LovableAPIs & webhooksMake or n8nVercel
Illustrative business uses, not additional modules or promised course projects. Actual use needs suitable data, permissions, testing and human review.
Add agents, retrieval and automation to your existing systems.
Prototype useful systems and turn problems into technical briefs.
Create practical work you can demonstrate and adapt.
Turn your process knowledge into connected workflows.
Arabic explanation. English materials and technical terminology.
No. Visual tools provide an entry point; coding topics are introduced with guidance.
At least 30, with a project in every session. The Projects section shows examples; the mix varies by group and not every build is deployed.
No. Industry examples show workplace applications, not promised curriculum deliverables.
A laptop you can install software on, a stable connection and relevant platform accounts. Some usage or hardware may involve extra cost.
The program includes a Space Code Academy completion certificate.
Your next move?
Bring a real task from your working day. Explore how to turn it into a focused, testable build.
What would you like to build first?
A smarter enquiry flow? An assistant for your documents? A useful internal tool?
Join the next AGENTIC group.
Ask Space Code Academy about the next start date, fees and registration.
Maadi, Cairo / 60 hours / 30+ projects
