Bella Baidak
Founder, developer, and product strategist. NSF and Cornell-backed. Building at the intersection of fashion, AI, and sustainability with the conviction that technology can fundamentally remake how the industry operates. Taste-driven technologist.
| Based | New York, NY |
| Domain | Fashion × Tech × AI |
| Engineering | React, Node, Python, SQL |
| Design & 3D | Rhino, Grasshopper, UX/UI |
| Research | Computer Vision, AI/ML, Privacy |
Bella Baidak
bellabaidak@gmail.com · New York, NY
Experience
Product Consultant
Jan 2026 – Present
Crossed
- ·Designed and launched a seller analytics platform consolidating sales data across Etsy, Vinted, PayPal, and eBay.
- ·Built Python-based integrations and automated data pipelines supporting cross-platform insights.
Product Lead & Co-founder
Jun 2023 – Oct 2025
Project B
- ·Defined product vision and roadmap for an AI-powered digital manufacturing platform translating body scans into production-ready garment specifications.
- ·Raised $685K through pre-seed funding and NSF SBIR grants.
- ·Led full product lifecycle from discovery and customer research through launch and deployment.
Product Research Manager
Oct 2021 – Present
Cornell University
- ·Led market research and opportunity assessment for privacy and AI products, informing product strategy and roadmap decisions.
Product Developer
Jan 2021 – Aug 2021
Broad Institute of MIT and Harvard
- ·Led redesign of a desktop scientific tool into a web-based platform supporting global biological image analysis workflows.
- ·Improved researcher workflow efficiency by 20% through iterative testing and product improvements.
Education
| 2021–23 | M.S Information Science, 4.0 GPA · Cornell University |
| 2017–20 | B.S Psychology, Minor in CS, 3.7 GPA · UMass Boston |
Awards
| $300k | NSF SBIR Phase I Grant, federal research award for technical innovation in fit-tech |
| $100k | Cornell Tech Startup Award, grant for early-stage venture development |
Project B
Developed software that translated body scan data directly into production-ready knitting instructions, powering a zero-waste on-demand manufacturing system. Built to replace mass production with custom-fit garments, starting with bras, where over 80% of women wear the wrong size.
| Type | Fashion Tech Startup |
| Role | Founder & Product Lead |
| Funding | $685k raised · NSF-backed |
| Press | Wall Street Journal |
| Impact | Zero-waste manufacturing system that produced hundreds of bras for women |
How It Works
01
On-phone 3D scanning
Depth sensor app that captures and stores an anonymous 3D scan. No photos or videos. Ever. Only raw depth data, processed into an anonymous 3D output.
02
AI algorithms & production-ready file generation
Once the scan is obtained it is analysed using feature mapping and volumetric analysis. A knit file is then generated using parametric inputs derived from the analysis.
03
On-demand 3D knitting
The generated file is automatically sent to the knitting machine and the garment is printed in one piece with zero waste in just 30 minutes.
Crossed
A platform for vintage resellers to centralize sales, inventory, and analytics across multiple marketplaces, making it easier to understand where, when, and how to sell most effectively. Built from years of firsthand experience as a vintage reseller, and developed in partnership with sellers across Europe and the United States.
| Type | SaaS Platform |
| Role | Founder & Developer |
| Status | Public launch July 2026 |
| Link | crossed.live |
Background
The resale ecosystem is deeply fragmented. Sellers are forced to manage listings, inventory, and performance across numerous platforms with no unified view. CROSSED addresses this directly, starting with a private client deployment and expanding to a public platform in July 2026.
Coming Soon
Cross-listing tools
List once, publish everywhere, across all major resale marketplaces simultaneously.
AI analysis
Intelligent insights on pricing, timing, and platform performance to maximise sell-through.
Gatekeep
A platform where users subscribe to creators' curated fashion finds and collections. Like Substack for fashion discovery, rewarding curation while encouraging more thoughtful and sustainable consumption.
| Type | Consumer Platform |
| Role | Co-founder & Developer |
| App | gatekeeep.com |
| TikTok | @gatekeep50 |
Preview
The Problem
Fashion creators earn income through affiliate commissions, which excludes vintage sellers, secondhand curators, and small brands entirely. Meanwhile, discovering unique secondhand pieces and emerging designers is fragmented and driven by algorithms, not taste.
The Solution
Gatekeep lets fashion creators monetize through subscriptions rather than commissions, opening up earning opportunities for curators who have always been excluded from traditional affiliate programs. For users, it makes sustainable shopping more accessible by surfacing unique finds through creators they already trust.
Publications
Custom Fit Bras From 3D Body Scans
SCF '23 · ACM Symposium on Computational Fabrication · Nov 2023
80% of women are wearing an incorrect bra size due to the current limitations in sizing, shape, and fitting methods of bras. Custom made bras could eliminate these fit issues, alleviating pain and discomfort while boosting women's confidence. We conducted a study using 3D scanning technology available on iPhones, where we had eight women take a scan of their breasts. From these scans, we extracted measurements and used them to generate custom digital bra patterns. Subsequently, we laser-cut the pattern pieces and sewed each woman a custom bra. Overall, women reported a superior fit with the custom bra but identified areas where further customization was needed.
↗ Read PaperCellProfiler Analyst Web (CPAW): Exploration, analysis, and classification of biological images on the web
IEEE Visualization Conference (VIS) · 2021
CellProfiler Analyst (CPA) has enabled the scientific research community to explore image-based data and classify complex biological phenotypes through an interactive user interface since its release in 2008. This paper describes CellProfiler Analyst Web (CPAW), a newly redesigned and web-based version of the software, allowing for greater accessibility, quicker setup, and facilitating a simple workflow for users. CPAW ports the core iteration loop of CPA to a pure server-less browser environment using modern web-development technologies, allowing computationally heavy activities, like machine learning, to occur without freezing the UI. We found that users could complete essential classification tasks with the same efficiency, and completed tasks 20% faster using CPAW compared to CPA 3.0.
↗ Read Paper| bellabaidak@gmail.com | |
| /in/bellabaidak | |
| Location | New York, NY |