Bridging Biology and Intelligence
We started Sryvo Twiya because there was this weird gap. Researchers had mountains of biological data but couldn't always extract the patterns hiding inside. Meanwhile, AI tools existed but weren't speaking the language of life sciences.
So we built something different. A team that actually understands both worlds—the computational frameworks and the biological questions that keep scientists up at night.
What Drives Our Work
Real Science
We're not interested in flashy demos. Our focus stays on building analysis pipelines that actually hold up when you're working with messy, real-world genomic datasets from clinical samples.
Honest Collaboration
Look, bioinformatics is complex. We don't pretend to have instant solutions. Instead, we work alongside researchers to understand their specific challenges and build tools that fit their workflow.
Taiwan Context
Operating from Kaohsiung gives us unique insight into the research landscape across Taiwan and broader Asia-Pacific networks. We understand local infrastructure and regional collaboration patterns.
How We Got Here
Early Research Phase
Back in 2021, we were just three computational biologists frustrated with existing tools. Spent months analyzing RNA-seq data for a cancer genomics project and kept hitting walls with available software packages.
Building Custom Solutions
By mid-2023, we'd developed our own pipeline for protein structure prediction tasks. Word spread through academic networks, and suddenly other labs wanted access to what we'd built.
Formal Establishment
Late 2024 marked our transition from informal collaboration to actual company. Set up operations in Kaohsiung's research district and started working with pharmaceutical partners on drug discovery applications.
Current Direction
Now we're focused on machine learning applications for metagenomic analysis. Projects scheduled through early 2026 involve microbiome research and environmental DNA sequencing collaborations.
How We Actually Work
- We start by understanding your specific research questions—not pushing pre-packaged solutions that might not fit your experimental design.
- Data quality assessment happens first. There's no point building sophisticated models on noisy datasets that need cleanup.
- Algorithm selection depends on your sample characteristics. We test multiple approaches rather than defaulting to whatever's trendy.
- Validation matters more than initial results. We build in cross-validation and robustness checks from the beginning.
- Documentation stays readable. Your future self (or your colleague) should understand what the pipeline does without decoding cryptic notes.
The People Behind the Analyses
We're a small group—computational biologists, data scientists, and research engineers who've spent years working at the intersection of biology and machine learning. No massive corporate structure, just focused expertise.
Liora Veldkamp
Lead computational biologist with background in transcriptomics. Spent five years at European research institutes before relocating to Taiwan. Handles algorithm development for genomic data analysis.
Joris Kyrklund
Machine learning engineer focused on protein structure prediction. Previously worked on neural network architectures for biological sequence analysis at pharmaceutical companies.
Research Infrastructure
Our Kaohsiung facility includes high-performance computing resources for large-scale genomic analyses and dedicated spaces for collaborative work with visiting researchers.