An ex-employee claimed that team members were routinely subjected to verbal abuse, even for basic clarifications or suggesting improvements.
The 'suicide' of a young techie of Krutrim, the artificial intelligence unit of Ola, has raised questions over the company's work culture after allegations of intense work pressure and verbal abuse against employees surfaced in a viral social media post.
Nikhil Somwanshi, who was a recent graduate of the Indian Institute of Science, Bangalore, and had joined Krutrim less than a year ago, allegedly died by suicide on May 8.
The incident gained widespread attention after a post by an anonymous user on social media platform Reddit went viral, describing the environment at the company as 'traumatic', especially for freshers.
The user claimed to be a colleague of Somwanshi.
"We are deeply saddened by the tragic passing of one of our most talented young employees, Nikhil, on May 8," an Ola Krutrim spokesperson told Business Standard.
"Our thoughts and heartfelt sympathies are with his family, friends, and loved ones during this incredibly difficult time," the spokesperson added.
The Reddit user in his post alleged that Somwanshi had been leading a project with two others despite being a recent graduate.
After the other two left the company, the workload allegedly fell entirely on him.
The post alleged that the team's manager was verbally abusive in meetings and disconnected from the team, working remotely from the US, while the rest of the staff was based in Bengaluru.
It alleged that the company authorities were attempting to suppress the news of the death.
An ex-employee, who previously worked at Krutrim, and collaborated closely with the team manager, corroborated the allegations and claimed to have resigned without another job offer in hand due to intense work pressure, recalling feeling suicidal.
Another former employee, who worked under the team's manager for several months, also backed the allegations, describing the experience as a "nightmare."
The ex-employee claimed that team members were routinely subjected to verbal abuse, even for basic clarifications or suggesting improvements.
Attempting to build with large language models (LLMs) was met with hostility rather than support.
The team was constantly under stress, and product output remained limited and subpar, the post alleged.
While the team manager worked remotely from the US, the former employee alleged he contributed little to actual execution and was known for giving unfocused speeches.
The company said Somwanshi was on personal leave at the time of the incident.
He initially reached out to his manager on April 8, expressing the need for rest, and was promptly granted time off.
On April 17, he informed the team that he was feeling better, but would benefit from additional rest, and his leave was extended accordingly.
"As a company, we are heartbroken by this loss. Nikhil was a valued team member, and his absence will be deeply felt by all who knew and worked with him," said the spokesperson.
"We are extending our full support to Nikhil's family and our employees during this time of grief.
"We are also in contact with the relevant authorities and will continue to offer our assistance as needed," the spokesperson added.
According to Somwanshi's LinkedIn profile, he worked on a large language model (LLM) and retrieval-augmented generation (RAG)-based chatbot called Saathi as part of his master's thesis, funded by the Melinda Gates Foundation.
The project was a collaboration between IISc Bangalore, Oxford Brookes University (UK), Kotak Mahindra Bank, and Akaike Technologies.
Saathi aimed to simplify access to over 200 government schemes across four Indian states by supporting regional dialects.
Somwanshi's key contributions included creating structured documents from scheme texts, developing a financial literacy module, and building systems for resolving user queries.
Feature Presentation: Ashish Narsale/Rediff
© 2025