I Didn’t Set Out to Build a Company. I Just Wanted to Build a Drug Discovery Engine.
- mansour ansari
- 2 days ago
- 3 min read
I Didn’t Set Out to Build a Company
So, here is my story. You see, when I started this journey, I wasn’t trying to build a startup. that was not in my mind. I am retired and live simple... no tension and no endless business meetings. I just wanted to code an engine with modern tools. I wasn’t chasing venture capital. I wasn’t thinking about pricing tiers, TAM, or business plans. Shoot! I am not even good at writing a business plan.
I had a much simpler and much harder goal:
I wanted to build my own cancer drug discovery system.
That’s it.
Why I Didn’t Buy Schrödinger (and couldn’t anyway) - Oh, man, that is 100K- to 200K software, and I am on Social Security! No way I could afford that!
Like many people interested in computational drug discovery, I was aware of the major commercial platforms. Schrödinger, MOE, OpenEye, the gold standards of the industry. They are powerful systems, built by brilliant teams. They are also completely out of reach for an individual. Six-figure licenses. Closed architectures. Vendor lock-in, etc.
Even if I could afford them, I realized something early on: learning to use someone else’s black box is not the same as understanding the physics and the process.
So instead of learning Schrödinger, I decided to learn what Schrödinger actually does and try to build its core myself. Not because it was easy. Because it was the only path available to me.
Building a Drug Discovery Engine on a Shoestring
I built my system on a couple of old PCs. No cluster. No enterprise software. No safety net.
I stitched together open tools, physics, mathematics, and a lot of patience. Along the way, I used AI to optimize things and provide direction. I learned docking deeply, not just how to run it, but how it fails, how it biases results, and how it explores chemical space.
Eventually, I had something that worked. Then I went further.
I containerized my docking system. I began pushing it toward the cloud, something most people never attempt with legacy scientific binaries. I built pipelines to sift through hundreds of thousands of compounds. I built ranking systems, rediscovery tests, and validation loops.
At some point, I realized I wasn’t just running simulations anymore.
I had built a drug discovery factory. I don't have anything better to do with my retirement time!
A digital factory that takes compounds in one end and produces short, ranked lists of plausible drug candidates on the other.
Quantum Curiosity Changed Everything
While working on this system, I was also studying quantum physics, initially out of pure curiosity. I wasn’t trying to “add quantum” to anything. I just wanted to understand randomness, uncertainty, and collapse at a deeper level. Then something unexpected happened. I noticed that many limitations in classical drug discovery pipelines weren’t just computational; they were exploratory. Classical systems tend to search the same comfortable regions of chemical space repeatedly, over-and over! Quantum entropy offered a different lever. Not magic. Not hype. Just a different way to inject uncertainty and diversity into search processes that desperately need it.
So I integrated it. Carefully. Empirically. Without assumptions.
That decision fundamentally changed how my system behaves, how it explores, how it ranks, and how it surfaces novelty.
I Still Wasn’t Trying to Build a Company
Even then, I wasn’t thinking about selling anything.
I wanted:
To build something real that is good for humanity in general.
To understand it end-to-end
To see if a single person, working patiently, could construct a modern drug discovery engine from first principles
Only later did people around me start asking questions:
“Why isn’t this a company?”
“Why aren’t you writing this up?”
“Why aren’t you thinking bigger?”
Blah Blah
That’s when business plans entered the picture, not as a starting point, but as a byproduct of building something that already worked.
Where I Am Now
Today, my system, now called QuantumCURE Pro™ — is a working, end-to-end drug discovery platform. It runs classical and quantum-enhanced workflows. It has passed rigorous benchmarks. It produces real results. It still runs lean, slowly but surely. It still runs on hardware I can afford. Ok, it takes time to dock 1000 compounds, but I sit and watch it! And I still treat it first and foremost as an engineering system, not a pitch deck.
Whether it becomes a company, a platform, or simply the most ambitious thing I’ve ever built, one thing remains true:
I didn’t start this to sell software. I started it because I wanted to build a cancer drug discovery engine — and I refused to accept that only large institutions were allowed to try.
Everything else came later.
FYI, yeah, I am building a Business plan and have a startup now!
Comments