If you are at an IIT, you already know that the internship season produces a particular kind of pressure. The large companies come early, the offers look impressive on a CV, and the social signal of a well-known name carries weight in a campus environment where everyone is watching what everyone else does.
But if you talk honestly to IIT graduates two or three years into those roles, a more complicated picture emerges. A lot of the work is narrower than it looked from the outside. The actual scope of what a junior engineer touches at a large company is smaller than what the brand implies. The learning curve flattens faster than expected, and the compounding — the thing that makes early career decisions matter — does not happen the way you assumed it would.
The question worth asking before you accept an offer is not “how does this look?” It is: “what will I actually be building, what will I be responsible for, and will I emerge from this three months from now materially more capable than I am today?”
That is the question this internship is designed to answer well.
Winvesta is a small, high-intensity AI-native fintech. We build two live products: a Global Currency Account for Indian exporters receiving payments from foreign clients, and an international investing platform for Indian retail investors. Both have real users, real complexity, and real technical infrastructure — payment rails, brokerage integrations, data pipelines, reconciliation systems — that needs to be built, maintained, and improved.
We are specifically looking for IIT engineers because we know what that selection process produces: people who can hold hard problems in their head, reason through them rigorously, and find a correct answer without being walked there. That is the foundation. What we build on top of it is what makes this internship different.
What We Offer That Is Genuinely Rare
You will own real work from day one. Not a scoped-down sandbox task designed to keep an intern busy. Work that affects real transactions for real clients, where your output either works or it does not and you find out quickly which.
You will use Claude Code as a primary engineering tool — not as a novelty, as the way work gets done here. You will learn its capabilities, its consistent failure modes, and the judgment required to be the intelligence in the loop between what it produces and what actually needs to ship. This is a skill that most IIT engineers your age are still figuring out how to develop. You will develop it here in weeks, not years.
You will work directly with the founding team. You will understand why architectural decisions get made, not just what they are. You will have the context that junior engineers at large companies spend years trying to acquire.
The last month of your internship is an explicit adjudication period for a full-time offer. You will know this from Day 1. Both sides deserve that clarity.
We look for the IIT selection itself as a signal of the mathematical and logical foundation we need — the ability to model systems, reason about correctness, and work through problems that do not have a clean answer in the back of a textbook. If you have made it through the JEE process and through your programme, that foundation is present. What we are looking for beyond that is what you have done with it.
Specifically: we want someone who has already started building with AI and has done so seriously. Not someone who has used Claude a few times, but someone who has used Claude Code on a real task, found where it went wrong, corrected it, and shipped something. The loop of delegate → evaluate → correct → ship is the core skill of an AI-native engineer, and it is the one we will develop most aggressively in this role.
We also need someone who is constitutionally skeptical of what AI produces. Claude is fast, capable, and wrong about something in every non-trivial engineering task. The engineer who ships what Claude Code produces without thinking about it is creating technical debt at speed. The engineer who evaluates every output, catches the error three layers from the surface, and fixes it before it compounds — that is the person we are looking for.
Beyond the technical, we need someone who moves without being asked. Who reads the situation, identifies the next step, and takes it. Who is genuinely excited about building in an AI-native environment — not treating it as a tool to make existing work faster, but as a fundamental change in what one engineer can produce alone.
Building and maintaining backend services and integrations that keep two live fintech products running — because in financial infrastructure, reliability is the product, and every failure has a consequence that reaches a real client.
Writing and improving data pipelines and reconciliation systems — because in a payments and brokerage business, the data has to be correct, and the engineer who ensures that is not doing maintenance work, they are doing product work.
Building internal automations and tooling using Claude — because a small, fast-moving team needs to eliminate manual work wherever it can be eliminated, and the engineer who builds that tooling multiplies the output of everyone around them.
Debugging and resolving issues that the team has not had bandwidth to address — because at our scale, problems do not get handed off to a dedicated oncall team. The engineer who can move fast through a problem they did not create, fix it, and document what happened is genuinely valuable.
Working across the full stack where the problem requires it — because in a company of our size, the distinction between backend engineer and data engineer and platform engineer is a constraint we cannot afford. You will work where the work is.
The specific shape of the work will change. That is expected.
We are not looking for a CV. We are looking for evidence of things you have actually built.
Link your GitHub if it has real work in it — not just coursework repositories. If you have used Claude Code on a real task, tell us what the task was, what it produced on the first attempt, and what you had to correct. If you have built a tool, a script, an automation, or a project that someone else used — tell us what happened after you built it.
The single most useful thing you can show us is the artifact of a real problem you solved and the decision you made about what to do with the solution. That tells us more than any description of skills.
This is an internship of 3–4 months minimum. Full-time conversion is possible and will be formally assessed in the final month — you will know this from Day 1.
You must be in Mumbai and able to come in to BKC every day. There are no remote options.
Your first week is dedicated entirely to Claude onboarding. Not codebase orientation, not sprint planning — Claude first.
Academic ideal: IIT, with strong performance in your programme. 90% or above in both 10th and 12th is what we look for. If your JEE rank and programme performance are strong, that is the signal that matters most. We care about the rigour the selection process represents, not just the institution name.
You must have used Claude Code on a real task before applying. Not a tutorial. A real problem, a real output, and at least one thing you had to catch and correct yourself. If that is not yet you, it will take you one afternoon to change that — and it is worth doing regardless of whether you apply here.
To apply: Use the portal where you found this listing. Please do not reach out to the team directly — applications submitted through the job portal are the ones that get reviewed.
If you are shortlisted, you will hear from us. We keep a close eye on responses and move quickly on the candidates we want to speak to. Due to the volume of applications we receive, we are not always able to provide individual feedback to those who are not selected — we regret that, and we mean it. It is a reflection of capacity, not of how seriously we take every person who applies.
Winvesta
Note: This is a paid internship.Skills: research and analysis,hubspot,communication,notion,content creation,team player,project management,team spirit,cross-functional collaboration,client engagement,proactive attitude,collaboration,proactivity,tech-savvy,market research,communication skills,analytical skills,data analysis
Dr. Reddy’s Laboratories Ltd. is a leading multinational pharmaceutical company based across global locations. Each of our 24,000 plus employees...
Apply For This JobLocation: BangaloreJob Type: Full-time Experience: Recruiting: 2 year (Preferred) Tasks 1. Talent Acquisition Strategy: Develop and execute effective recruitment strategies...
Apply For This JobOur Purpose Mastercard powers economies and empowers people in 200+ countries and territories worldwide. Together with our customers, we’re helping...
Apply For This JobThis role has been designed as ‘Hybrid’ with an expectation that you will work on average 2 days per week...
Apply For This JobBasic Details: Fill the required information about business, unit, location, position, reports to position and date of updation of JD...
Apply For This JobWe are seeking a committed and results-driven Executive / Senior Executive, Marketing to support the planning and implementation of strategic...
Apply For This Job