I am a computer science student at Iowa State University who likes turning complicated systems into clear, useful products. My work moves across developer tools, native Apple software, applied AI, and security research, but the standard stays the same: understand the system, make deliberate tradeoffs, and leave behind something people can trust.
- Building Parable, an open-source library of accessible React components and complete application templates.
- Developing Parcel, a native macOS capture and editing studio built around one shared rendering pipeline.
- Exploring private verification and collective storytelling through Mosaic, a SwiftUI and Supabase project.
- Publishing technical field notes about software architecture, applied ML, and adversarial thinking.
| Project | What it is | My focus |
|---|---|---|
| Parable | An open-source marketplace of accessible React components and multi-framework templates. | Component architecture, distribution, documentation, and product design. |
| Parcel | A native macOS capture studio using SwiftUI, ScreenCaptureKit, Vision, and a shared render pipeline. | Native interaction design, media pipelines, and on-device workflows. |
| Mosaic | A shared artwork where privately verified acts of kindness become equal-weight visual tiles. | Product concept, SwiftUI, Supabase, privacy boundaries, and visual systems. |
| Formwork | A repository of production-minded application templates and interface patterns. | Reusable systems, information architecture, and developer experience. |
| FaceCheck Pro | A deliberately limited educational face-retrieval architecture study. | SCRFD detection, ArcFace embeddings, vector retrieval, and explicit safety constraints. |
| BBHUNTER | An AI-assisted security research pipeline with phased reconnaissance and independent validation. | Research orchestration, evidence quality, and adversarial validation. |
Product & developer tools → Parable · Formwork
Native software → Parcel · Mosaic
Applied AI → FaceCheck Pro · image and sentiment studies
Security research → BBHUNTER · technical field notes
I do not treat these as disconnected specialties. Product work teaches me to care about clarity; native development sharpens attention to interaction and performance; applied AI forces careful boundary-setting; security research makes assumptions visible. Each discipline improves the others.
- Building a face search engine end to end — Aug 2026
- How I do recon — Aug 2026
- The model that finds the bug will defend the bug — Aug 2026
- Start with the real constraint. Architecture is better when privacy, accessibility, performance, and failure modes are treated as inputs rather than cleanup.
- Prefer evidence over spectacle. I value reproducible results, honest limitations, and documentation that makes decisions inspectable.
- Design the whole path. A strong implementation includes the interface, the system behind it, and the explanation that helps another person extend it.
- Keep learning in public. Projects, case studies, and writing are working records—not claims of perfection.
You can explore the complete case studies on my portfolio, read longer notes on Shellcat, or connect with me on LinkedIn.
Ames, Iowa · biswas06@iastate.edu