A naukri.com initiative
Yourstory
1M
390
Image Credit: Yourstory
How realfast.ai is helping IT services industry cut delivery times
- realfast.ai, a Singapore and Bengaluru-based startup leveraging AI to accelerate execution, aims to double revenue and halve delivery times for IT companies to address challenges faced by the Indian IT services Industry.
- By deploying AI models for testing, development, deployment and maintenance, AI agents help human workers speed up the delivery process while maintaining or even improving the code quality and service standards.
- Vayu, the company's core offering, which is built on top of multiple AI models such as ChatGPT and Anthropic’s Claude, has improved software systems by enhancing traditional and outdated code, making them more efficient and delivering better results
- Vayu which helps Salesforce implementation teams work faster, has already seen early success, with its unit-testing agent reducing the time for writing unit tests by 50%, from a typical 10-day sprint cycle to just eight days
- Although the AI sector in India is expected to reach $17bn by 2027, the market remains largely untapped in the IT industry for which every role, from sales to QA, deals with text-based tasks such as contracts, requirements, code, logs and tickets, it is an area that is often ignored due to its complexities.
- realfast.ai's AI platform allows AI agents to operate on top of deep, complex enterprise stacks, boosting efficiency across key service domains (including ERP, ITSM and HCM implementations).
- After a year in stealth mode, realfast.ai has raised seed funding from PeakXV, RPG Global and DeVC Dugout and is currently in early pilots with mid-sized IT companies in India.
- Ponnappa also stresses that raw LLMs trained on academic benchmarks lack the consistency required for enterprise work, stressing the need for task-specific agents that deliver better results.
- One of the key trends that we foresee in the industry with AI adoption is that human beings will now spend a lot more time reviewing the work than generating.
- The main challenges come from concerns about IP (intellectual property) and privacy, though there has been progress in addressing these issues, says Ponnappa.
Read Full Article
23 Likes
For uninterrupted reading, download the app