A · The problem
Why Skysurf exists
If you have ever bought a stock on someone's recommendation and watched it fall, you already know the problem. You don't know why they recommended it. You don't know why they would tell you to sell, or when. You don't know what they did differently for the client who paid more. You are trusting a process you have never been allowed to see.
Traditional PMS in India starts at fifty lakh of capital. The shape of the firm explains the floor: analysts making calls, custodians holding the money, fees that pay for both. For most people who have built up real money in equities through their day job, that has never been an option. We built Skysurf for that gap.
B · The structure
How Skysurf is built differently
Skysurf is a systematic equity strategy. A fixed set of rules produces every decision; nobody on our side is paid to have an opinion. Your shares stay in your demat account, so there is nothing to custody. The communication runs on software, so there is no relationship manager and no quarterly printed report.
This is not a workaround. It works because broker APIs in India are now good enough to place trades in your account while you keep full control of your money. The minimum can come down because the work that justified the old minimum can now be done by a system instead of a room full of people.
C · The method
The methodology, in plain terms
The strategy runs once a week on weekly market data. It scans roughly 1,000 NSE-listed stocks above a market-cap floor, picks the setups that match its entry rules, ranks them, and places the trades through the broker's API. Each position is a flat one percent of equity. Thirty positions at most; no more than five from any single sector.
Exits are handled by trailing stops that ratchet up as a position moves into profit, then tighten once it crosses defined thresholds. There is no discretion in the live loop. The same input produces the same output, every time. The rules do change, but only when a new version has survived walk-forward testing against years of data it has never seen.
The full code, the validation notebooks, and the configuration that produces the live strategy are available on GitHub.
D · The evidence
Performance
What the strategy has done
Walk-forward means the strategy was tested on years it had never seen during construction.
Over that period Skysurf returned 24.2% per year, against 9.6% for the Nifty 50. Skysurf's worst peak-to-trough decline was 12.3%; the Nifty's was 34.6%. Annual return divided by maximum drawdown gives 1.96 for Skysurf and 0.28 for the Nifty, roughly seven times the risk-adjusted return.
How the strategy behaved during specific market events
Most of the people who lost real money in Indian equities over the last decade lost it in one of three windows: the 2018 small-cap correction, the COVID crash of 2020, and the 2022 sector rotation. The chart above shows how Skysurf moved through each one. Two of the three went well: in COVID and in 2022, Skysurf finished comfortably ahead of the index. The 2018 slowdown did not. Skysurf ended that year 3.0% below where it started; the Nifty ended 2.9% above. We fell deeper and recovered slower.
Skysurf is not designed to avoid every loss. No honest strategy is. It is designed to keep the losses smaller and the recoveries shorter, and most of the time it does.
Drawdowns
This is what holding Skysurf through every losing patch of the last sixteen years would have felt like. The worst point is 12.3% below the previous high. The longest stretch underwater is 26 months, from November 2010 through December 2012. Most drawdowns resolved within six months.
Look at this chart at least as carefully as you look at the equity curve. The equity curve is the destination. The drawdowns are the journey.
Annual returns
Skysurf beat the Nifty in most years, but not every year. In some years a buy-and-hold investor in the index would have done better, and that is the cost of a defensive, systematic approach. Across the full period the compound return is well above the index; the year-by-year picture is bumpier.
What did not work
What is on this page is the result of years of testing, most of which produced ideas that did not survive. A few are worth naming.
Tighter position sizing on higher-conviction setups looked better in-sample but did not hold up in walk-forward. Every position is still a flat one percent.
Time-based exits, closing positions after a fixed holding period, kept clipping our winners. There is no time stop.
US-style momentum frameworks ported straight to India produced worse results than building from Indian price history alone. So we built it on Indian data from the start.
Late in validation we caught one of our ranking signals using a forward-looking calculation we had missed. Once we fixed it, headline returns dropped by about a third. The numbers on this page are the corrected version. We are naming this because it is exactly the kind of error that is easy to commit and hard to spot, and we would rather you see the correction than not know about it.
Known limitations
A few caveats worth being explicit about.
The validation universe is built from stocks listed today, not the universe as it existed at each point in time. Survivorship bias probably inflates the historical returns by 1 to 3 percentage points a year.
Sector tags are applied retroactively using the 2026 mapping. For most stocks the difference is small; for some it is not.
Walk-forward starts in 2010, so the 2008 financial crisis is not in the out-of-sample window. The full-period numbers do include it.
E · The terms
What Skysurf will and will not do
The strategy is disciplined, not frozen. Execution is mechanical and never overridden in live trading. The rules do evolve, but only once a change has survived walk-forward testing against historical data.
The cadence is weekly. Skysurf does not generate daily action. We built it for people who have other things to do.
The fee is a flat annual subscription. No performance fees. We do not earn more when the strategy trades more. The actual number is tiered by capital and shared in the conversation.
Your capital stays in your demat account. Skysurf has no way to move your money out of it. The strategy places trades; your broker holds your shares.
There is no stock of the day. Skysurf is not a tip service, a Telegram channel, or a recommendation feed. The strategy generates a portfolio. The portfolio is what it is.
The methodology is open. The code is on GitHub. The validation is on GitHub. If you want to audit the strategy before you put money in, you can.
F · The people
Who is behind this
Skysurf is built by two people.
Akhilesh designed the strategy. Twelve years of personal investing in Indian equities, three of them full-time on this. Background in product. He treks.
Manish built the engineering: the systems that run the strategy each week, talk to broker APIs, and keep everything working. Without him this is a research notebook. He rides motorcycles.
If you have questions about how Skysurf was built, or why we made specific choices, we would rather answer them in a conversation than try to anticipate them on a webpage.
G · The first step
How participation begins
You have read this far. If Skysurf is the kind of thing you want to be part of, the next step is a conversation. Join the waitlist below; we admit a small number of investors each month and set up a call within a few days.
I read every message myself and respond within a few days. If you would prefer to write directly, the address is akhileshkhajuria.work@gmail.com.