In recent times, Generative AI (GenAI) has grown in leaps and strides, discovering purposes in several fields, particularly software program growth. Integrating AI and GenAI into the software program growth lifecycle comes with quite a few advantages, enhancing each part from idea to deployment. From automating repetitive duties to streamlining workflows and lowering errors, these applied sciences are redefining the software program growth course of.
Nonetheless, the emergence of latest instruments on daily basis and the seemingly limitless variety of use instances could be overwhelming for builders to navigate the GenAI panorama. At DevSparks Bengaluru 2025, Rahul Bhattacharya, AI Chief, EY GDS Consulting, explored how builders and engineers can apply AI-powered instruments to remodel enterprise purposes in real-world tech environments.
Listed here are the highlights from the session.
Constructing a problem-first mindset
The huge variety of AI-powered predictive and generative instruments accessible to builders at the moment is each a boon – and a bane. The sheer selection is usually overwhelming, and discovering the precise instrument is usually a problem. Nonetheless, Bhattacharya recommends discovering the issue earlier than selecting the instrument. “Issues,” he mentioned, “exist unbiased of options. Search for the precise instrument to use to unravel the issue. So, at all times begin from the top aim and work backwards to seek out the instrument.”
GenAI instruments: reward or threat?
The dialogue additionally tackled the moral makes use of and dangers of leveraging these instruments. Whereas Bhattacharya notes that moral concerns could not essentially apply to coding, he highlighted the danger of utilizing copyrighted code in coaching massive language fashions (LLMs).
Organizations must cope with the dangers of choosing AI/ML instruments. “Take into consideration your urge for food for threat inside the enterprise. What’s the risk that, even when there’s a 1% probability, what could be the price of that consequence? As a result of threat isn’t just a likelihood. It is usually the affect of one thing going incorrect. Should you bear in mind likelihood and statistics, you’ll acknowledge the idea of anticipated worth (the likelihood weighted worth of any random occasion). So, it is vital to have a look at that and resolve: Is it definitely worth the threat?” he mentioned. Threat requires a case-by-case analysis.
He cited the business aviation trade as a robust analogy. Regardless of 80% of in-flight operations being automated by way of superior management techniques, sensors, and actuators, the stakes are so excessive that each flight nonetheless requires two pilots to supervise the journey. “Yearly there are a dozen papers concerning the feasibility of autonomous aviation, however we’ve not even gone from two people to 1 human within the cockpit,” Bhattacharya famous. At this level, he believes threat analysis is essential in relation to leveraging AI/ML instruments. Organizations should rigorously assess the extent of oversight and human management required in an AI-driven system and proceed accordingly.
Driving innovation for customer support
Bhattacharya additionally mentioned how EY GDS – an innovation hub and expertise powerhouse inside the world EY group – gives enterprise options to member corporations and is making use of AI/ML instruments to deal with shopper challenges and drive innovation. He defined how AI-powered options on the world EY group are being utilized throughout the enterprise – from the entrance workplace to backend operations and the center workplace. The entrance workplace, which generally handles buyer care, customer support, operations, and gross sales, requires a distinct set of options.
Take the instance of contact facilities. “When prospects name in, they usually hear that the dialog will probably be recorded for high quality assurance functions. In actuality, solely 5% of these calls are literally reviewed because it’s simply too useful resource intensive for human groups,” Bhattacharya mentioned. The worldwide EY group addressed this by integrating GenAI instruments that now consider 100% of buyer calls.
From these outcomes, organizations can create customized coaching for particular person brokers, sharing vital suggestions to enhance the client expertise. Moreover, contact facilities sometimes spend quarter-hour after every name logging notes on buyer satisfaction, follow-ups, and many others. With GenAI instruments, these duties could be streamlined with automated summaries generated, reviewed, and submitted with minimal handbook effort.
Streamlining contracts for center places of work
Center places of work deal with provide chain, stock, processing techniques, contracts and different departments that sometimes contain heavy paperwork, mentioned Bhattacharya. He pointed to massive telecommunication corporations that operate by way of contracts and contract administration.
To deal with inefficiencies and scale back threat, the worldwide EY group launched AI-powered instruments that may extract key entities, test for compliance to requirements, and spot variations throughout a number of contracts, thereby enhancing the worth of the agreements, lowering dangers, and assist allow compliance.
Erasing the drudgery of knowledge lineage
The again workplace, which homes the builders, IT system administration, and extra, is ripe for disruption with GenAI and AI instruments. One of many frequent tasks the again workplace handles is cloud migration, significantly the shift from on-premise infrastructure to the cloud. The group acknowledged that on this course of, establishing knowledge lineage is extraordinarily vital. Sometimes, this job has been handbook and liable to errors: workers learn by way of the scripts, hint knowledge sources and file it in a spreadsheet, after which repeat the method advert nauseum. To streamline this course of, the group launched GenAI techniques in order that the primary spherical of this course of is automated.
“Quite a lot of the work we do at EY GDS is enterprise transformation tasks. Knowledge migration is a big a part of the method. Discovering the info lineage, and doing supply to focus on mapping? Not attention-grabbing or technically difficult. It is drudgery. All of this may be automated and make us extra productive by taking away the low value-add work,” he shared.
EY GDS has made it such that builders don’t even want to jot down a immediate to automate these processes – the immediate is engineered and positioned inside the IDE.
Bhattacharya mentioned a spread of use instances for predictive and GenAI instruments, together with one in promoting. Through the use of multimodal fashions and laptop imaginative and prescient, the worldwide EY group was capable of scale back a course of from 45 minutes to a mere 45 seconds, with all model tips within the advert being adopted. He concluded the session with a name to builders to embrace AI, put money into studying, achieve expertise, earn related certifications and take the leap into India’s AI-driven workforce.
<div class="externalHtml embed" contenteditable="false" data-val="”>
Elevate your perspective with NextTech Information, the place innovation meets perception.
Uncover the newest breakthroughs, get unique updates, and join with a world community of future-focused thinkers.
Unlock tomorrow’s developments at the moment: learn extra, subscribe to our e-newsletter, and develop into a part of the NextTech neighborhood at NextTech-news.com
