Empty folder on Monday. Something people can open by Friday.
Everyone has a project they have started. Almost nobody has one they have finished, and a half-built thing proves nothing to anybody. This program is about finishing: choosing something small enough to actually ship, using AI to move fast through the boring parts, reading what it hands back rather than trusting it, and getting the thing onto a URL a stranger can open. You bring the idea. You leave with it live.
- Duration
- 5 weeks · around 20 hours
- Mode
- Live online — real sessions, not recordings
- Timing
- Evenings, designed around a full-time job
- Cohort size
- 20–25
- Reviews
- Your code read by an engineer, twice
- Recordings
- Every session recorded, yours to keep
Who this is for
- Students with a GitHub full of projects that stop at eighty percent
- Career-switchers who need one real thing to point at, not five tutorials
- Working developers who want to see how far AI actually moves the needle
- Founders who want to build the first version themselves before hiring anyone
Anyone hoping to type a sentence and receive a finished product. AI collapses the typing, not the thinking — you will still make every decision here.
Prerequisites
- Comfort with a code editor and the basics of any one language.
You do not need to be good. You need to be able to run a file and read an error message. If you have finished Claude at Work or AI Foundations, you are more than ready.
Modules
Choosing something you can actually finish
The single biggest reason projects die is that they were too big on day one. We cut your idea down until it fits five weeks, write the one sentence it has to satisfy, and agree what is deliberately not in it. Then the repository, the README and the first commit — before any feature exists.
IN THE SESSION
Cut your idea down until it fits five weeks, write the one sentence it must satisfy, and open the repository with a README before any code exists.
You leave with: A scoped idea, a repository, and a README written before the code.
Blank page to running skeleton
Getting from nothing to something that runs is where AI earns its keep. Scaffolding, wiring, the boilerplate nobody enjoys. You will also meet the failure mode early: code that looks right, runs, and does the wrong thing — and how to catch it in minutes rather than on Friday.
IN THE SESSION
Get from an empty folder to something that runs and does one real thing, then plant a bug in it yourself and see whether you can spot it in the generated code.
You leave with: A skeleton that runs on your machine and does one real thing.
Making it real
Data that persists, state that survives a refresh, and the unglamorous half: what happens on bad input, on a failed request, on an empty list. This is the difference between a demo and something you would let a stranger use.
IN THE SESSION
Make it survive a real user: persistence, bad input, a failed request and an empty state, each tested by the person sitting next to you.
You leave with: A project that behaves when someone uses it wrong.
Reading what the machine wrote
Line-by-line review of your own codebase, including the parts you did not write yourself. Where it invented an API that does not exist, where it duplicated logic, where it silently swallowed an error. Then your first real code review, from an engineer, on your actual project.
IN THE SESSION
Read your own codebase line by line, mark every part you cannot explain, and take a written review from an engineer on it.
You leave with: A codebase you can explain, and a written review of it.
Ship it
Deploy, a live URL, and the README that gets read — screenshot at the top, one paragraph on what it does, one line on how to run it. Then everyone demos, and we review what shipped.
IN THE SESSION
Deploy, get the live URL, write the README that gets read, and demo it to the cohort.
You leave with: A live URL with your name on it, and your Codnov Certificate of Completion.
What you walk away with
- One finished project, deployed, that a stranger can open
- The habit of reading generated code instead of trusting it
- Two written code reviews of your own work by an engineer
- A README and a repository you would be happy to link on a CV
- Codnov Certificate of Completion, with a unique ID and a public verification link
Sample projects
- A tool that solves one problem you personally have every week
- An assistant over a body of documents you already own
- A small internal tool for a team you are part of
- A public site that does something useful and boring, and does it properly
You finish with a deployed, reviewed AI project, or we re-run you free.
Requires minimum 75% attendance and on-time milestone submission. Full conditions at codnov.ai/training/terms.
Taught by Codnov engineers who build and run AI systems in production every day.
Cohort pricing is confirmed per intake — enquire for the current fee and dates. All prices exclusive of 18% GST.
Nothing is paid on this website. Register your interest and we will confirm your seat with you directly.
Register your interest
We will come back to you with the full syllabus, the schedule, and whether there is a seat in the next cohort.
Codnov Certificate of Completion. Not a degree or diploma recognised under the UGC Act, 1956 or the AICTE Act, 1987. All prices exclusive of 18% GST. Project completion guarantee subject to minimum 75% attendance and on-time milestone submission; full terms at codnov.ai/training/terms. Open to participants aged 16 and above.