Research at JustKalm
Our commitment to scientific rigor and transparency. Explore our whitepapers, model cards, and research publications that underpin every algorithm and decision.
Research Areas
Our research spans multiple disciplines, combining expertise from academia, industry, and regulatory bodies to advance the science of product intelligence.
Valuation Intelligence
12 publications
Multi-signal market valuation using Bayesian methods, ensemble ML, and real-time data fusion.
Research Lead
Dr. Jennifer Chen
Academic Partners
Health & Toxicology
8 publications
Computational toxicology, dermal exposure modeling, and skin microbiome impact assessment.
Research Lead
Dr. Sarah Kim
Academic Partners
Sustainability Science
15 publications
Lifecycle assessment, circular economy metrics, and environmental impact quantification.
Research Lead
Dr. Marcus Williams
Academic Partners
Responsible AI
6 publications
Bias detection, fairness metrics, model interpretability, and ethical AI deployment.
Research Lead
Dr. Priya Patel
Academic Partners
Behavioral Science
5 publications
Nudge design, decision architecture, and sustainable behavior change interventions.
Research Lead
Dr. Alex Rodriguez
Academic Partners
Data Science & ML
10 publications
Production ML systems, uncertainty quantification, and real-time inference at scale.
Research Lead
Dr. Kevin Liu
Academic Partners
Model Cards
Transparent documentation of our ML models following Google's Model Cards framework
Valuation Engine
v2.4.0 · Updated 2024-12-01
Multi-signal product valuation model combining market data, material analysis, and brand metrics.
Accuracy
98.6%
vs 91.2% baseline
MAPE
1.43%
industry avg 4.8%
P95 Latency
45ms
SLA: 100ms
Coverage
99.2%
50K+ products
Intended Use
Consumer product valuation for resale platforms, insurance, and retail analytics.
Known Limitations
- •Performance may degrade for ultra-luxury items (>$10K)
- •Vintage items require additional provenance signals
- •Limited coverage for regional/indie brands
Out of Scope
Sustainability Scorer
v1.8.0 · Updated 2024-11-15
Multi-dimensional sustainability scoring integrating LCA, certifications, and supply chain data.
Correlation w/ LCA
0.94
vs manual review
Certification Accuracy
99.8%
200+ certifications
Coverage
45K+ brands
Update Frequency
Daily
Intended Use
Product sustainability assessment for consumers, retailers, and brands.
Known Limitations
- •Relies on brand-disclosed data for supply chain metrics
- •Emerging certifications may have delayed coverage
- •Regional regulatory variations may affect scores
Out of Scope
Health Risk Assessor
v1.5.0 · Updated 2024-11-20
Computational toxicology model for dermal exposure risk assessment from textile chemicals.
Chemical Coverage
500+
ZDHC MRSL + REACH
Regulatory Alignment
99.5%
vs ECHA database
EDC Detection
98.2%
confirmed disruptors
Response Time
< 50ms
Intended Use
Health risk screening for product development and consumer guidance.
Known Limitations
- •Cannot detect chemicals not in database
- •Exposure modeling assumes standard wear conditions
- •Individual sensitivity not accounted for
Out of Scope
Publications
Multi-Signal Product Valuation: A Bayesian Approach to Consumer Goods Intelligence
J. Chen, A. Rodriguez, M. Williams, S. Kim
We present a multi-signal valuation framework that combines real-time market data, material composition analysis, brand positioning, and sustainability metrics to produce accurate fair market value estimates. Our Bayesian approach achieves 98.6% accuracy (MAPE 1.43%) across 50,000+ validation samples, outperforming single-signal baselines by 23%.
Quantifying Sustainability: A Multi-Dimensional Scoring Framework for Consumer Products
M. Williams, E. Chen, P. Anderson
This paper introduces a comprehensive sustainability scoring methodology that integrates lifecycle assessment (LCA), supply chain transparency, certification validation, and end-of-life recyclability. We validate our approach against 12 existing frameworks including Higg Index, B Corp, and EU PEF.
Computational Toxicology for Consumer Products: Dermal Exposure Modeling and Risk Assessment
S. Kim, R. Martinez, L. Wong
We develop a computational toxicology engine that estimates dermal exposure risk from textile chemicals using established toxicokinetic models (Potts-Guy equation). Our system integrates REACH, ZDHC MRSL, OEKO-TEX, and EPA ToxCast databases to provide comprehensive health risk assessments.
Model Card: JustKalm Valuation Engine v2.4
JustKalm ML Team
Comprehensive model card documenting the JustKalm Valuation Engine including training data, evaluation metrics, intended use, limitations, and ethical considerations following Google's Model Cards for Model Reporting framework.
Textile-Skin Microbiome Interactions: A Computational Framework for Health Impact Assessment
L. Wong, S. Kim, A. Patel
This paper presents a novel framework for assessing the impact of textile materials on skin microbiome health. We model fiber-microbiome interactions, antimicrobial effects, and skin site-specific sensitivity to provide actionable health guidance.
Measuring Circularity: End-of-Life and Material Recovery Metrics for Fashion
E. Chen, M. Williams, J. Park
We introduce a comprehensive circularity scoring system that evaluates take-back programs, material recovery rates, recycling infrastructure compatibility, and waste diversion potential. Our framework aligns with Ellen MacArthur Foundation Circulytics.
PFAS Detection in Consumer Textiles: Regulatory Compliance and Health Risk Modeling
R. Martinez, S. Kim, L. Wong
This technical report details our approach to PFAS (per- and polyfluoroalkyl substances) detection and risk assessment in consumer textiles, including bioaccumulation modeling and regulatory framework mapping for REACH, TSCA, and Proposition 65.
Uncertainty Quantification in AI-Driven Valuations: A Monte Carlo Framework
A. Rodriguez, J. Chen, P. Anderson
We present a comprehensive uncertainty quantification framework that combines Monte Carlo simulation, ensemble methods, and Bayesian credible intervals to provide confidence bounds for all predictions. Our approach enables transparent risk communication for high-stakes decisions.
Case Study: AI-Powered Sustainability Scoring at Scale - Patagonia Partnership
JustKalm Partnerships Team
A detailed case study of our partnership with Patagonia, demonstrating how AI-powered sustainability scoring enables transparent consumer communication and drives purchasing decisions toward more sustainable options.
Scientific Advisory Board
Our research is guided by world-class experts from leading academic institutions and research organizations.
Dr. Emily Osborne
Scientific Advisor, Toxicology
Harvard T.H. Chan School of Public Health
Environmental toxicology, endocrine disruptors
Dr. Michael Chen
Scientific Advisor, Sustainability
Stanford Woods Institute for the Environment
Lifecycle assessment, circular economy
Dr. Sarah Nakamura
Scientific Advisor, AI Ethics
MIT Media Lab
Responsible AI, algorithmic fairness
Dr. James Thompson
Scientific Advisor, Behavioral Science
Duke Center for Advanced Hindsight
Behavioral economics, decision architecture
Dr. Lisa Martinez
Scientific Advisor, Dermatology
UCSF Dermatology
Skin microbiome, contact dermatitis
Our Open Science Commitment
We believe transparency drives trust. All our model cards, methodology documentation, and key research findings are publicly available. We actively engage with the academic community and welcome peer review of our methods.
Research Collaboration
Interested in collaborating on research? We partner with academic institutions, industry labs, and NGOs on sustainability, health, and AI ethics research.
Contact Research TeamCiting Our Research
To cite JustKalm research in academic publications, please use the following format:
Available at: https://justkalm.com/research/[paper-id]