With a keen interest in AI/LLM I recently build out a website analysing 6 County data originating from the Food Standards Agency API https://www.food.gov.uk/ - this is under Open Government License https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/
The FSA operates a rating scheme where public health officials visit food establishments and rate them. In the North this comprises almost 17,000 establishments - the good, the bad and the ugly.
Please visit my https://publicfoodratings.com/
Technology Stack
- Python & Flask
- Backend API and web framework
- Google Cloud Platform
- BigQuery, Cloud Run, Vertex AI
- BigQuery
- Data warehouse and analytics
- JavaScript & Bootstrap
- Frontend and responsive design
- Vertex AI
- Machine learning and predictions
- Enterprise Security
- Google Cloud security standards
The rise of the LLM and its foray into coding has created interesting opportunities for those who wish to build out something in an area that they find of interest.
Although the Integrated Development Environment has been around a while we now see the incorporation or twinning with code aware Large Language Models. This may enhance developer productivity through features like code generation, autocompletion, debugging and codebase understanding. On the other hand it may only lead to bad code but certainly to more competing ideologies.
In this case I used an IDE called Cursor - https://cursor.com/ in combination with Claude, Gemini and ChatGPT.
Each of the LLM have their own idiosyncrasies, e.g.
- for navigating / deploying Google Cloud I believe that Gemini knows its way through the labyrinth
- to temper the impetuousness of Cursor you could use Claude to issue instructions in very controlled terms (and make backups)
For the above graphic I asked ChatGPT to produce an abstract composite of my tech stack using a vibrant palette
