When Akanksha woke up in her flat in Bengaluru around 7 am on March 31, the first thing she did was check her email. This had become a routine ever since the 24-year-old, who requested that only her first name be used, started working with a US-based team at Oracle. But instead of the usual exchanges, there was an email from the company saying there would be an important update. Within an hour, her access to the company Slack channel was cut. When she reached out to her manager, she was told that she, along with 40% of her team, were being let go of.
“I didn’t think I’d be laid off – I was the only developer working on a feature that was important for the client,” says Akanksha, who hails from Jharkhand and had joined the tech company three years ago straight out of IIT Roorkee.
Around the same time, Akhilesh Sharma, who was working remotely for Oracle from his hometown in Himachal, was trying unsuccessfully to log in to start his work. Assuming it might be an issue with his VPN, he checked his inbox, only to find an e-mail similar to what Akanksha had received. It was when he reached out to his manager that the picture became clear – that he was being laid off.
“Ironically, I woke up that day thinking that this is a good company and I should stay for 4-5 years,” says the 28-year-old software engineer.
Akanksha and Akhilesh are among the 10,000 to 12,000 Oracle employees in India and an estimated 30,000 globally who lost their jobs on March 31. The US-headquartered company, which had over 1,60,000 employees as of May 2025, has not officially commented on the job cuts. But in early March, Bloomberg reported that the company, headed by tech mogul Larry Ellison, was planning to cut thousands of jobs as part of efforts to deal with the cash crunch brought on by its huge investments in artificial intelligence (AI) data centres. At least some of these cuts would be in roles the company expects it will need less of due to AI, according to the report.
In India, the retrenchment package includes two months’ salary, an additional 15 days’ pay for every year at the company, and Rs 20,000 as medical insurance premium, apart from gratuity, etc.
But Akhilesh said the bigger blow was losing out on encashing the Rs 10 lakh of employee stock options (Esops) he held, since he was being let go before completing a year. In the time he was there, some others in his team had already been laid off, so this was not entirely unsurprising. But it was still unsettling.
“I was there only for 8 months so it affects me less. But I heard of someone who was laid off after being with Oracle for 17 years. The layoffs may make sense from a business point of view, but think about it from a humanitarian point,” he says on the phone, a few days after the announcement.
The Oracle layoffs stand out for the scale – but they are also part of a series of job cuts at tech companies, globally and in India, that have been taking place over the last couple of years and are at least partly tied to the inexorable march of AI.
On Thursday, April 23, Meta announced plans to cut 10% of its workforce globally, translating to about 8,000 employees, and close another 6,000 open roles as the social media giant ramps up investments in AI. The same day, Microsoft announced its first ever voluntary buyouts for employees in the US.
Closer home, TCS, India’s largest IT services firm, had announced in mid-2025 that it was shrinking its workforce by 2% that financial year, translating to about 12,200 jobs. Its peer Infosys terminated the contract of hundreds of trainees, a move which attracted the scrutiny of Karnataka’s Labour Department. One report estimated that India’s top five IT companies, the flagbearers of its $300-billion IT outsourcing industry, saw net hiring – the number of employee additions minus the number let go of – of just 17 in the first nine months of FY26. This is a sliver of the almost 18,000 net hires in the same period a year ago. More recently, Wipro stated during its fourth quarter FY26 results that it does not have a fresher hiring target for the next financial year, compared with its recruitment of 7,500 new employees from campuses a year ago.
“Will AI replace me” has become both a talking point and source of concern ever since Open AI’s viral release of ChatGPT in November 2022, giving rise to terms like ‘jobscalypse’ and ‘AI-mageddon’. Three years later, some of the talk around entire professions getting replaced appears to have been more hype than reality – at least as of now. And the huge swathes of job cuts in tech cannot be attributed to AI alone. But it’s also becoming increasingly evident that the tech and IT sector in India, which used to hire lakhs of graduates and was considered an aspirational and ‘safe’ white-collar route to financial security, is in the midst of a structural transformation, with employees bearing the brunt.
For techies, popular shorthand for software engineers, it’s a time of increasing anxiety and uncertainty. Some talk of the fear of being replaced, the inevitability of layoffs, and the need to look at alternative sources of income. Young engineering graduates worry about the dip in recruitment at the entry level and the kind of jobs they will be able to find. Analysts warn of the lack of skilled recruits to match the new roles being created.
Giving more cause for concern is the fact that this is unfolding in an economy that already struggles to generate enough jobs and where 67% of unemployed youth (between 20 and 29 years) are graduates. All of this signals the brewing of a potential crisis, even as governments at the union and state level vie to court the multi-billion dollar data centre projects that power AI models and services.
“The existential crisis hits me every day”
Kailash Raj* was interested in coding right from his school days, finding it fun and logical. With the added prospect of a stable career that would pay well, he went on to study computer science engineering, joining the India office of an American cloud computing company after his master’s. Kailash’s firm recently began giving employees licences to use Cursor, an AI-powered coding assistant, to use at work.
The difference, he says, is stark. For instance, writing test code used to be an important but time-consuming and dull part of his job – but with Cursor, something that would take days now takes mere minutes, sometimes just seconds. “The amount of time we save in repetitive and basic level tasks is significant,” he says.
But even as it has made his job easier, Kailash, who’s now a senior engineer, is also struggling to come to terms with how quickly AI is doing something that used to take him days. “It makes me feel less important, less mission-critical – and kind of useless,” he says.
The existential crisis, he says, hits him every day. Half his team was laid off last year, with no explanation. The thought that he might face the same fate any day is constantly on his mind.
Akhilesh, the engineer who was at Oracle, also describes himself as a geek who was into software since high school and was certain that computer science engineering was what he wanted to do. But once he heard of ChatGPT, he says he started realising that basic coding would be rendered useless over the years. The recent layoff has left him with even more questions.
“Just the other day, my manager was talking about how he created a workflow using an AI agent that was able to raise a peer review request for a code change (to ensure code quality) on its own and close that ticket. It got me thinking about what is even our point in the organisation – the engineer’s responsibility is changing.”
Thinking about the job market, he says, makes him anxious.
Anxiety about layoffs in the tech sector by itself is not new – with AI adoption, there’s an added layer of uncertainty, with many professionals worried not just about the short term but about whether their roles will remain relevant in the long term, says Anisha Kumar, a clinical psychologist and researcher with Kaha Labs, a research arm of mental health organisation Kaha Mind.
“What stands out is that this anxiety often appears even among individuals who are currently employed and performing well,” she says.
Richa Singh, CEO and co-founder of online counselling startup YourDost, says questions around how AI will impact their jobs and whether they will be made redundant are concerns the platform’s counsellors are hearing more frequently in the last year from clients in tech and other sectors. “It’s also something we often hear from students who are just graduating. With entry-level roles, a common sentiment is ‘everything can be automated’.”
AI layoffs or AI washing?
India’s IT services companies had about 8 million employees in 2023, according to a report by the government thinktank NITI Aayog. The report, titled ‘Roadmap for Job Creation in the AI Economy’, estimated that this could reduce to 6 million by 2031 due to the risk of AI.
In April, the number of job openings in tech fell to 1,10,000 from 1,19,000 the previous month and was less than half the 2,56,000 active job openings in April 2022, according to IT staffing firm Xpheno. Its founder, Kamal Karanth, is however loath to ascribe all of this to AI alone, particularly with macro conditions like the war in West Asia impacting the global economy.
“At the moment, it’s dicey to say this is AI-driven. IT spending has come down and companies are correcting for the over-hiring they did in 2022,” he says.
Researchers from Ashoka University’s Isaac Centre for Public Policy (employment and labour vertical), based on their ongoing work, note that between 2018–19 and 2025, data from the Periodic Labour Force Survey shows that India’s employment growth has been concentrated at both ends of the AI-exposure spectrum.
“Jobs least exposed to AI – such as mining and construction labour, cleaning, and food preparation – have grown the fastest on average, while highly exposed roles in software, finance, and sales have also seen some expansion, though relatively less than the former. In contrast, occupations with moderate exposure have seen little to no change,” they said, in an email to TNM.
Even as layoffs in tech are gathering speed globally in the last couple of years, distinguishing how many of these are due to AI has been a challenge globally as well. Tech layoffs trackerlayoffs.fyi estimates that in the first four months of 2026 alone, over 73,000 tech workers have been laid off from 95 tech companies, compared to over 1,25,000 in all of 2025.
Tech honchos have spoken about how AI is dramatically changing the way we work and figures like Anthropic CEO Dario Amodei predict the end of software engineering by 2027. But multiple experts, including influential Silicon Valley figures like investorMarc Andreessen, have expressed the view that AI is also being used as a fig leaf to correct pandemic overstaffing. This practice of linking job cuts to AI and productivity gains instead of admitting to overhiring even has a term – ‘AI washing’.
But there is more consensus on the impact of AI on entry-level jobs in software. In late 2025, an analysis of millions of job openings by King’s College London from 2021 to 2025 found that UK firms whose workforces are highly exposed to AI capabilities reduced total employment by 4.5% on average, “with the effect concentrated almost entirely in junior positions, which fell by 5.8%”. Another analysis by researchers at Stanford’s Digital Economy Lab says there is early evidence consistent with generative AI disproportionately impacting entry-level workers in the US labour market. Employment for young software developers was down by nearly 20% in September 2025 compared to late 2022, when ChatGPT was rolled out, it found.
In India, data shows that IT companies are now hiring fewer graduates. According to Quess Corp, the country’s largest staffing company, the top 10 IT services companies alone hired 2.5 lakh to 2.75 lakh fresh graduates in 2022 and 2023. That has nearly halved in the last two years, to 1.25-1.5 lakh.
Kapil Joshi, CEO of Quess IT Staffing, says there are a couple of reasons for this shift: one, companies are looking for hires who are “billable” immediately, in other words, those with experience who can begin work straight away, without training. Second, the productivity gains due to AI, which has automated certain tasks like testing, means fewer people are needed for the same job. “Earlier, there was huge demand for testers. But with automation, you need fewer testers. There is less demand for certain tasks like infrastructure support, typically part of entry-level roles,” says Kapil.
It’s a reality playing out across IT firms, whether in Bengaluru or Belagavi. Take the experience of US-returned tech entrepreneur Hitesh Dharmdasani, who has been deploying AI agents for the last year-and-a-half at the network security products startup he runs from his hometown, Belagavi.
“You can’t assign something with a broad scope to an AI agent. But it can do small tasks well, which is what freshers usually do,” he says.
He explains that if he had to build a small proxy server, for instance, which would need about 20,000 lines of code, a senior engineer would typically break that down into 40 features split up among 10 junior engineers. Now, those 40 small tasks can be assigned to five AI agents, which can do it more efficiently.
“This is the situation everywhere – an experienced engineer with an agent is 5-10x more productive than if they were to hire someone and bring them up to speed.”
For companies, using agents is also more cost-effective than paying for the salaries of entry-level hires who need to be trained at least for a few months. As a result, firms like his and a few others in the city which together used to hire 200-300 graduates in a year now hire about a third of that, says Hitesh, who is also a founding member of the Belagavi Technology Companies Association.
For engineering students like Aarya Gaikwad, who is in the final year of her B Tech in computer science in Belagavi, this is bad news. “I took computer science because I heard it had good scope for the future. But now I’m a bit scared about the job market,” says the 22-year-old, who is now trying to build her skills in AI.
Her batchmate, Aavishkar Yallurkar, has got one offer but since it’s for a non-technical role in AI data labelling at an annual salary of Rs 3 lakh, he is looking for other options. “When I joined the course in 2022, the final year batch of 60 got 120 offers,” he says.
Quess’s Kapil says there is more to the dip in entry-level hiring than a lack of jobs. “Roughly 1.5 million engineering students graduate every year – of this only 15% understand emerging technology and are job-ready. This 15% would be from the Tier 1 colleges that would have invested in updating the curriculum. Tier 2, Tier 3 colleges would be struggling with placements.”
His scepticism on the role of AI in layoffs aside, Xpheno’s Kamal agrees that mass hiring at the entry level in IT will no longer be a reality. “We have to find a way to rechannelise this – it’s a big ask for India.”
Uncertainty is the new certainty
In discussing how to deal with this rapidly evolving job market, one term that often comes up is ‘upskilling’, whether you’re a fresh graduate or a more experienced employee. Both Aarya and Aavishkar, the final year engineering students, mention that they are doing courses on their own, to be more ready for the new job market.
Akanksha, the former Oracle employee, plans to explore more AI-related jobs. “Every time the tech industry goes through a churn, it comes up with new jobs. The traditional way of working may change but software is not going anywhere – I have to adapt as fast as possible,” she says.
Akhilesh feels that with AI making coding easier, it may create more entrepreneurs and multiple small companies. With so much uncertainty, though, he says he would also explore options for a secondary source of income. “I feel AI will be good for society, but personally, I’m worried,” he says. He is also considering helping build an AI tool that others could use, drawing a comparison with the industrial revolution. “The main beneficiaries then were those who built the machines and those who used them.”
Quess’s Kapil strikes an optimistic note about the impact of AI on tech jobs. “My personal view is that it will create more jobs, make us more productive and increase our per capita income.” This would, of course, depend on having an adequately trained workforce. “India missed the manufacturing bus. It can’t afford to miss the AI bus.”
Still, even with upskilling and AI creating new roles, the experience of software companies already shows that automation of certain tasks, if not entire roles, would mean fewer people are eventually needed for the same job, which is bound to create some short-term pain. Many companies, especially global tech giants, have also announced large outlays for building data centres, which require investments of billions of dollars. If these companies need to cut costs elsewhere to accommodate this switch, layoffs could be one way – as was seen with Oracle.
IT unions are wary of how this will all play out. “AI is making our jobs easier. But our concern is that, if 100 people were needed for things like testing earlier and AI reduces that to 50, what happens to the other 50 employees? These are people who pay taxes, who have EMIs to pay,” says Vinod AJ, general secretary, All India Forum for IT/ITES Employees (AIFITE), the oldest union for IT employees. “Companies are talking about upskilling but not everyone will be retained.”
The IT sector, he says, employs 80-90 lakh people directly. “We need to figure out how to address this. There has to be a proper support system. Right now, employees are voiceless,” he says. AIFITE is in discussions with other unions and there are plans to make a representation to the government, he adds.
Kailash, the senior engineer working in the multinational tech firm, says his coping mechanism for now has been upskilling himself in AI development. “I asked Claude to build a syllabus for me. It’s the only way forward if you don’t want to get wiped out.”
Even then, the thought of being replaced by AI is never far. “Unless you’re a co-founder or an entrepreneur who’s building something, AI will eventually do your work as a backend developer – and by eventually, I mean four-five years,” he says.
* Name changed to protect identity
Indulekha Aravind is an independent journalist based in Bengaluru.