Hey everyone — it’s been almost four years since I last wrote a blog post. Once in four years feels like the Olympics, except with fewer medals.

Over these years, I’ve run several “A/B tests” on myself.

Except, if we’re being precise, they weren’t A/B tests at all. They were pre-post dramas where you change something, forget you had a baseline, and then congratulate yourself when anything at all happens right.

Test 1: Extreme Diets

Goal: To see if a radical eating plan could completely reboot my energy and appearance.

Two versions of me went in. Version A ate like a South Indian with insulin resistance, Version B ate like a saint preparing for sainthood. 6 months later, Version B was 15 kg lighter, my mornings felt like they belonged to a superhero, and, curiously, social warmth from certain women dropped by what felt like 60%.

The diet achieved 100% of its energy goals, 70% of its aesthetic goals, and -50% of its social harmony goals. Unfortunately, there was later interference from Test 3 — the back break leave — during which I ate with the kind of reckless joy that brought me quietly back to Version A.

Edit:

Some of you asked how I managed to lose 15 kg. The truth is, I had help — from Amura.ai, founded by Saravanan Kumar. I first heard of Amura back in 2020, when not one but two CEOs I’d worked with underwent extreme body transformations purely through diet. At the time, Amura was a quiet little secret; now it’s out in the open, thanks to Vidya Balan’s interview.

If you ever want to try it, I can connect you. I owe my transformation to Saravanan and his team — and their rare mix of science, precision, and the kind of persistence that politely ignores all your excuses.

Test 2: Name Change on LinkedIn

Goal: To learn if a different name could change my networking luck.

I became “Dana” for a while, curious to see if non-Indian ethnicities would click “Connect” more readily. Connection requests from the US and Europe rose by about 42%, The net effect was a small but noticeable increase in cross-cultural conversations, which, I suspect, was worth the minor confusion in my own circles.

I’m running an experiment on my LinkedIn for a few weeks where my name sounds more American/ Non-Indian (Dana G) and want to see how Request Acceptance looks like

With my original name it was OK – Made 250 cold connections across 3 weeks.

— Vandana Guru (@vandana_guru) July 4, 2025

Test 3: The Back Break Leave

Goal: To restore physical health without losing professional rhythm.

I stepped away briefly when my back began issuing daily protest notes. The break worked — pain dropped to near zero — but my inbox grew as if it had been breeding.

It also overlapped with a surge in indulgent eating, which sabotaged Test 1 by quietly reversing some of the gains from Version B. In the scorecard of my life experiments, this was a rare case of back testing — revisiting old conditions, with results that confirmed I am still perfectly capable of undoing my own progress.

I’m back to work, and now breaking backs (also called alignment)

Test 4: Accent Swap (Multi-Factorial Crossover Test)

Goal: To measure how speech patterns — and even vocal identity — affect how much I’m heard.

This began simply: switching between a neutral North American cadence and my natural Indian- English accent. Then I layered in an AI twist using Krisp AI, which transformed my voice to sound more like a man.

In A/B testing language, this was a multi-factorial crossover test — running two variables (name from Test 2 and AI-modified voice) in parallel on the same social network.

It was revealing in ways I didn’t expect. The AI’s idea of making me sound confident was changing my voice to sound like a man.

Amusingly, if I practiced this in non-social life as a woman, I’d be described as “abrasive” rather than “highly competent.”

The funniest part was nobody questioned. Even when I came clean mid-conversation and said the voice was AI-generated, most people shrugged and carried on.

Eventually, I ended the trial, put Krisp AI away, and went back to being myself — Vandana Guru, a name meaning “the teacher who is worthy of reverence.”

These days I speak in a semi-neutral tone that’s all mine, neither algorithmically adjusted nor borrowed from another gender, and somehow still entirely fit for purpose.

Tesla’s Door Handle Design: My Favourite Pre-Post Amusement This Month

Recently, Tesla’s flush door handles caught my attention. They’re sleek, theatrical, and unmistakably Tesla.

I imagine the goal was straightforward: improve aerodynamics and brand signature without making the driver feel they’re operating machinery from the 1990s.

In execution, the handles likely shaved up to 12% off the drag coefficient, extended range slightly, and gave every new Tesla owner a two-second “aha” moment the first time they popped out. That’s about 90% success in design impact, 70% in aerodynamic gain, and maybe 60% in all-weather practicality.

The alternatives — traditional fixed handles, push-button latches, or phone-triggered pop-outs — might have scored higher on ease of use but far lower on drama. Tesla clearly optimised for theatre and efficiency, not familiarity.

Next Personal Test

Inspired by Tesla, my next pre-post experiment will be to open my Uber Tesla door in two styles:

• Version A: Walk toward the handle with calm authority and let it pop out to greet me.
• Version B: Use only my pinky finger to flick it open, like a secret handshake.

My prediction: Version A will score high on dignity and reliability, but Version B has strong potential for a 42% lift in passenger amusement and a 20% boost in my own private joy. And, true to my method, the final verdict will depend entirely on whether I feel like keeping score.