Ved Antani is the SVP of Engineering and Managing Director, India at New Relic. He brings over two decades of leadership experience across global tech companies, including Twilio, JioCinema, BetterPlace, and Myntra. Ved has played a pivotal role in scaling engineering teams, driving platform innovation, and leading large-scale digital transformations. He has also built and led high-impact R&D centers and product organizations across India and globally. Ved currently leads the talent expansion and operations for New Relic in India
Intelligent observability has become a critical and necessary tool in any business’ arsenal now, not just for improve operations, but also for driving better results on the bottom line
By Ved Antani, SVP – Engineering and Managing Director, India at New Relic
We’re in the age of AI. Its speed and efficiency make it attractive to businesses that want to scale, personalise, and improve customer experiences. Today, AI is the core strategic priority for 80% of businesses in India, with many investing in cutting-edge technology to stay ahead of the game. However understanding how the AI is truly performing presents a key challenge for many organisations, especially when business resilience relies on the AI’s ability to take over repetitive work and predict failures before they occur. These knowledge gaps can result in unplanned downtime which can cost businesses dearly. The 2024 Observability Forecast found that 65% of businesses in India have experienced downtime with the cost averaging at $1 million USD per hour. From disrupted services to lost revenue, the real cost isn’t just the outage, it’s also the hours of untapped productivity. This is where intelligent observability coupled with agentic orchestration and AI-Powered predictions can help.
Agentic orchestration for faster, more efficient outcomes
Agentic orchestration is the ability for systems to talk to each other to help resolve issues. Intelligent observability tools integrated with powerful agentic orchestration, identify issues before they arise, assess impacts, and make intelligent recommendations. It works by leveraging a compound AI system that uses multiple AI models, agents, and tools to tackle many types of complicated tasks. Agentic AI, meanwhile, reduces the developer toil by completing tasks and automating workflows, even with external tools.
With traditional monitoring, alerts were triggered only after a problem occurred or a predefined threshold was breached. It would put site reliability engineers into a reactive mode as they tried to minimise the impact on their users. Agentic orchestration is changing this for the better. AI can now predict outcomes by leveraging machine learning (ML) algorithms to analyze historical data, identify patterns, and predict time-series metrics in a unified space. Businesses can conduct ad hoc predictions on a given data set or as an alert condition that looks ahead for early warning signs of a problem.
For example, a manufacturing unit operates heavy machinery, such as industrial presses or conveyor belts. Downtime halts production, costing millions. To avoid production delays and unplanned downtime, typically, maintenance teams reacted only once failure occurred. The business still had to bear the cost of downtime even if the issue was resolved quickly.
Intelligent observability with agentic orchestration is preventive. It gathers historical data from various machines and sensors, such as temperature, pressure, vibration, and energy consumption, and learns over time. By applying machine learning algorithms to this data, the manufacturer can detect subtle changes in patterns before a failure actually occurs. Maintenance teams are notified early, and preventive maintenance tasks can be scheduled ahead of time, minimizing unplanned downtime. Predictive analytics enables teams to reallocate resources to perform repairs when production is not at peak, avoiding disruptions to the manufacturing schedule. The power of preventive observability is unparalleled. It can put businesses on the path of zero downtime.
Lesser incidents, greater profitability
Software developers, regardless of the industry they work in, rely on frequent code changes to innovate quickly. Code changes are also the leading cause of software incidents. AI Agents are making it easier than ever to solve this problem. AI Agents built into intelligent observability tools evaluate changes across the digital ecosystem on their own. Multiple AI Agents work together to detect issues from code changes and address them directly in the integrated development environment. Such real-time response eliminates time-consuming manual processes and reduces the risk of code changes. As a result, businesses can boost productivity and improve software quality.
For instance, in retail and ecommerce, if the business is facing an issue with the inventory management system, intelligent observability solutions don’t just flag the problem. They immediately alert the right team, whether that’s IT or operations staff, and provide intelligent recommendations on how to solve the issue so they know exactly what to do to fix it. If there’s a software bug in the inventory management system, AI automatically and proactively identifies it, and escalates it to developers via GitHub, without requiring manual intervention. It even generates code fixes via Code Assist to make life easier.
As businesses adopt AI on a large scale, they need intelligent tools to monitor their health and ensure their performance is smooth. Intelligent observability is now a critical and necessary tool in any business’ arsenal, not just to improve operations, but to drive better results on the bottom line. With Agentic AI taking center stage, the issues of downtime don’t have to hamper efforts to become more profitable. If AI is the future of business, agentic orchestration is the future of observability, and it is already here.
