hey, i am damir. you probably came here from one of three things.
the research. biological metrics get used to decide which tumour cells are likely to survive treatment, and the standard way of validating them cannot separate a score that is measuring biology from one that is counting how deeply the cell got sequenced. so i built the separation out of linear algebra over rank residuals and the topology of protein interaction graphs, calibrated it on 39,505 cells whose developmental order was fixed under a microscope before sequencing ever happened, and audited thirteen of them. five of the seven i could measure survived. the code is open source. the long version
the company. pragma builds ai automation infrastructure: agent systems, data pipelines, and parsers that turn scattered documents and crm records into something you can query. open models, on hardware the client controls, so their data never leaves the building. pragma.com.kz
the writing. young founders, titles, and the difference between having one and solving something. essays
before all of it: vetra ai, built alone, with an offer that valued it at $8m. kuramoto, sold, and my first exit. 400+ engineers and founders mentored at nfactorial. upraze, built so kids would stop paying prep agencies thousands. ten years of taekwondo and the kazakhstan national team. work