2024-03-26 12:19:00
The use of generative artificial intelligence (GenKI) is already having a significant impact on software development. Many companies that use these tools report significant increases in productivity. One According to a McKinsey study With the help of GenKI, software development can be accelerated by a factor of two. In particular, providers of application software and database systems have jumped on this bandwagon and are offering appropriate support.
Advertisement
While Oracle presented the generation of SQL queries using natural language input last fall, Salesforce followed up with an extensive portfolio tailored to developers at the recent TrailblazerDX (Salesforce Developer Conference). This is where voice input contributes to code generation Anypoint Code Builder, the workflow tool Salesforce Flow and the programming language Apex are used. According to the company, for example, business processes can be formulated in natural language, which are then mapped fully automatically in Salesforce Flow. Apex code should also be able to be created in the same way. The voice input is also supported by auto-completion, so that the wording can be done more quickly.
The generation is not limited to application code: the associated test scenarios can also be created automatically with a click of the mouse. They are usually not perfect right away, but are intended to serve as a starting point with which, according to Salesforce, any test environment can be set up much more quickly than if developers start from scratch. “We’re not interested in making developers and administrators superfluous; instead, we want to relieve them and speed up their work,” emphasized Alice Steinglass, Executive Vice President and General Manager of the Salesforce platform, in an interview with iX.
Bridging the gap between the IT and business worlds
In this context, Steinglass sees another advantage of code generation using natural language: “We know that there is still a gap between the business and IT worlds. With AI, this gap can finally be bridged, because if I “I’m able to formulate a workflow or a business process in natural language, then both sides understand it and IT can continue working directly with the generated code,” the Salesforce manager is convinced.
She illustrates this using two examples. The first involves a mailing, for which a series of data must be retrieved from the CRM system in order to then use it in an appropriate advertising form for a highly personalized email. The approach requires a combination of business expertise and IT/data knowledge. This means you need someone who understands the business and knows exactly what you want to achieve with this email. And then it requires someone who knows the data and knows how to retrieve it. GenKI can be the bridge through which both sides can reach a concrete agreement.
The second example involves formulas, some of which can be very complicated. To this end, Salesforce has expanded its Einstein software – including Einstein 1 Studio, which was also presented at TrailblazerDX and helps integrate AI into Salesforce applications – to include the formulation of formulas in natural language. This means that employees from different specialist areas can “speak” the formulas from which new code is then created. Developers then have to check this generated code and refine it if necessary, for example to better reflect exceptions. Here too, GenKI serves as a bridging element to improve communication between business and development.
Silvio Saravese, Executive Vice President and Chief AI Scientist at Salesforce, goes one step further: “Natural language offers the ideal opportunity to lower the inhibition threshold for using IT tools. That means it’s not just about programming, but Every tool and every complex software – everything can be used much more easily when used by voice.”
An important step, but still far from perfect
But not everything is perfect yet – on the contrary. Oracle admits that its SQL code generation is only 70 to 75 percent accurate. The Salesforce managers also emphasize that “developers definitely have to look at the generated code very carefully.” The extent to which code generation can be further perfected is uncertain. However, experts believe that a significant error rate will remain because natural language is not clear enough – especially when it comes to nested queries with many exceptions.
Salesforce is therefore not focusing so much on perfection in future development, but rather on designing new roles. Saravese has his own prediction: “Developers will become designers. This means that the AI will take over many of the manual tasks, while developers concentrate on fine-tuning and putting together individual modules – just as a composer determines the instrumentation of his work.”
However, this does not make programming superfluous. Saravese even explicitly recommends that schools continue to teach programming. “Programming knowledge will continue to be essential in the future, because as long as you can’t trust the AI, every automatically created code has to be checked manually – and that’s only possible with in-depth knowledge,” he explains his assessment. Apart from that, he is convinced that learning a programming language is still the best training for logical thinking.
(map)
To home page
1711458762
#code #generation #Salesforce