My very first post. I am excited to be here.
I retired from the software industry years ago and now work from a small back office in Oklahoma City. Most people think retirement means stepping away from building things. I saw it differently. So, about fifteen months ago, I started a project that eventually became QuantumCURE Pro™. (https://citizenscientist.org/) The goal was ambitious but simple: build a cloud-native platform that could help accelerate early-stage drug discovery before anything ever reaches a wet lab.
What started as curiosity quickly became an obsession.
I spent thousands of hours buried in Python/JS code, cloud architecture, scientific papers (online), Docker containers, databases, APIs, and computational chemistry workflows. The deeper I looked, the more I realized that many research tools were powerful/super expensive, individually, but fragmented when combined into a real workflow. A total mess of workflow.
I wanted to see if a single engineer could assemble a practical, multi-user scientific platform using modern cloud infrastructure.
Today, QuantumCURE Pro™ orchestrates molecular docking campaigns directly from a web browser:
Researchers to Central Queue to AutoDock Vina to Results & Golden List
Results & Golden List, various reportings and forensic trail, to (MD, GANN, to Skala) to Wet list ....
**
So, **researchers upload compound libraries, jobs enter a queue, workers scale on demand, and results flow back through dashboards, mobile monitoring, APIs, and validation tools.
The surprising part wasn't getting AutoDock Vina running.
The hard part was everything around it:
*** Multi-user orchestration
- Worker scaling
- Queue management
- Reliability tracking
- Mobile monitoring
- Cloud deployment
- Data provenance
- Dashboard truthfulness, and I can list more.**
One of the biggest milestones came when I finally got Physics-based AutoDock Vina running reliably inside a cloud container. That took me months of work. That was the moment I realized the real project wasn't docking/scoring, it was everything around docking.
Once the engine could run consistently in the cloud, I could start building worker pools, queues, APIs, monitoring, scheduling, mobile access, and multi-user workflows around it. The problem shifted from "Can I run this scientific software?" to "Can I scale it into a reliable platform that real researchers can use?"
So, the journey continued toward perfection and trust building. After more than 5,000 hours of solo development, the challenge is no longer building features. The challenge is proving reliability, documenting behavior, and preparing the platform for real researchers. I will get there!.
This is my first post on DEV. In future articles, I'll share what I learned about worker pools, cloud orchestration, scientific software, dashboard design, and building a complex platform as a solo founder at an age when most people expect you to stop building altogether.
Ask me a question about the workflow. I am here to answer.
Top comments (0)