A product designer embedded with client product teams — marketplace operations at Meesho, student support at Vedantu, cardiac digital therapeutics at Lupin, industrial waste management at Cirqlar.
The work I am proudest of is usually the work that changed direction. Two of the case studies below are about a brief that was wrong. One is about a brief that was right but silent on the thing that mattered most. One is about a brief that had to be reframed before a line of it could be built.
Sellers believed they were losing half their returns. They were losing 3%. That belief was sitting inside Meesho's prices.
The 6-minute walk test is done with a clinician watching, because heart failure patients can get into trouble during it. We had to move it into people's homes.
A quarter of respondents said they couldn't find support. Seven interviews later, we had cancelled the redesign that was about to fix it. The output was a funded initiative stopped and redirected.
Five factories reported to global sustainability frameworks on top of numbers hand-logged in paper challans. The brief asked for a reporting layer. The work was digitising capture on the shop floor.
I would rather show you evidence with its limits attached than a confident number you cannot check.
Read a case study ↗In each case study you will find the sample size, who ran the research, and what I inherited versus what I generated.
The Meesho retest was one participant who had already been asked the same question three times, and I say so.
At Lupin I had no patient access at all, so I went to a cardiologist, and I am precise about what that substituted for and what it did not.
Eleven years at Auriga IT, embedded with clients. Someone will check LinkedIn, and a discrepancy costs more than the clarification does.
Meesho, Vedantu and Lupin were client engagements, not employment. They are labelled that way everywhere on this site.
As an embedded designer I am usually scoped to delivery. The measurement window opens after the engagement closes. Where I could chase numbers afterwards I have, and where I could not, I say so rather than presenting targets as results.
Since 2023 I have been on Auriga's own products. Cirqlar was the exception — a full B2B platform I led end-to-end, now live as SaaS. Alongside it, design systems work, Figma tooling and AI-assisted design workflows. Less of the deep product problem-solving I want to get back to, which is why I am looking now.
Execution is table stakes at this level. What I bring is a habit of interrogating the brief before I commit a team's cycles to it, and enough discipline about evidence to know when my interrogation is well-founded and when it is not.