Doctoral Researcher, Gene Therapy Laboratory, VIT Vellore, India · ahmedaneesm@gmail.com
Synthetic biology is getting easier to design, but harder to trust.
I'm a gene therapy researcher who ships software. I build the infrastructure required for the next era of biological design: automation, lab memory, and a reliable co-scientist, bridging wet-lab reality and dry-lab scale.
At the bench: genome engineering, viral vectors, mammalian cell culture, CRISPR, integration-site analysis.
In code: end-to-end agentic workflows, machine learning models, and quantum methods for genomics.
AI agents can handle the execution. But judgment about what to build, and whether a design actually survives the physical reality of a living cell, cannot be delegated. That's the part I do. What I build, I also test.
Most of what's here is research-grade tooling for genome engineering and computational genomics. These five are the ones worth your time.
- pen-stack - a verification and grounding layer for AI-driven genome writing: models propose edits, PEN-STACK checks each one against validated tools, never a language model.
- bio-firewall - genome-writing-native biosecurity middleware that inspects the plan a design agent produced, not just the final sequence, across five hazard axes.
- discern - a coupled disease-variant engine for differential diagnosis, misdiagnosis prevention and VUS resolution in inherited bleeding and platelet disorders.
- janus-design - joint amino-acid and codon design for de novo proteins: the sequence and the DNA that encodes it, optimized together.
- quasar - quantum algorithms for mutation-selection dynamics, with a straight assessment of where the quantum-classical boundary actually sits.