One project within contemporary AI is to enable everyone to programme with nothing other than English prose.
DeepSeek comes close to achieving this goal.
This is how far I got with roughly 2 hours of DeepSeek Coder. For background, I only have a basic grasp of the broad web development landscape, and only one prior project with the technology involved (Python Flask).
My first 8 prompts were as follows:
- Hong Kong litigators: create a Python Flask Calendar programme to help lawyers calculating filing dates for documents.
- Default the date to today.
- Display the calendar as an image, like most calendar apps. Display both the Start Date and the eventual Date on the calendar
- When I run locally “Access to 127.0.0.1 was denied You don’t have authorization to view this page. HTTP ERROR 403”
- What to put in requirements.txt
- [Pasting in an error message]
- Calendar does not reload when I enter a new date
- Handle situation of going to the next month or even the month after
Up to this point, everything just worked.
Hiccups arose, leading to some blind ends and the copy and pasting of error messages. But DeepSeek Coder more or less did the following for me out of the box:
- Style the calendar
- Make sure the date is calculated backwards from the hearing date.
- Add the rules of calculation of dates i.e. 72 and 48 hours before the hearing (excluding Saturdays, Sundays and general holidays).
- Debug the situation where the hearing date (surprisingly) falls on a Saturday or Sunday
- Add CSS styling.
- Deploying for free on Render
Revisiting it the following morning, I made some very minor changes. I am confident of making meaningful further improvements if I spend more time on it.
Arguably, this is not quite no code programming. I did have to know what “Python Flask” and “CSS” are. But the bar has lowered dramatically.
Is this a positive development? Only time will tell. Despite all the free services around the internet, prior to LLMs, it was still possible to maintain the illusion that our programming skills are somehow innate to us.
It was nurtured, no doubt, with Google and Stack Overflow, but nevertheless it was the fruit of our own labour.
Is a project like this the fruit of my own labour?
There is a saying that when some web service is free, you are the product. Suppose a programmer starts learning and spends her entire learning journey improving her skills with various LLMs. Is she a user of LLMs? Are (the people who build and fund) LLMs using her?