Founder
Priyanca Ford
Founder & Chief Pepper Officer, Scoville Splice LLC
We started with two acres and a question.
In 2014 we got a piece of land and started growing peppers — just as a hobby, just because we liked them. I grew up at the intersection of Indian and Japanese food cultures, where heat is a real ingredient rather than a stunt, and peppers had always felt like something worth understanding properly.
By 2018 we had a greenhouse. That summer we made fermented hot sauces and spent weeks feeding them to the family — testing heat levels, talking about the difference between front-of-mouth and back-of-throat burn, cataloging what worked. It was around that time the question sharpened: what actually limits how hot a pepper can get?
Every answer led to more questions. We kept growing plants. I started feeding my years of genomics work into the biology of the capsaicinoid pathway — and what came out of that was the Capsaicin Design Platform.
The computing part I did myself.
I ran RNA-seq on Capsicum chinense placental tissue at five developmental time points, quantified 28,074 genes, and identified 4,365 that were differentially expressed during peak capsaicinoid production. Then I built a 37-metabolite, 44-reaction constraint-based metabolic model of the full pathway in COBRApy and ran flux balance analysis to find out which step actually limits output.
The key finding
Capsaicin synthase (Pun1/AT3) — the enzyme everyone assumed was the rate-limiting step — came back with a flux control coefficient of zero. The vanillylamine branch carries 90% of pathway flux control. Engineer those four upstream genes, and the pathway runs. That single result changed the entire engineering strategy.
From there: ESM2 protein language model to scan 7,920 possible mutations in capsaicin synthase, AlphaFold2 to validate the top candidates, five CRISPR-SpCas9 guides targeting the peroxidase enzymes that degrade capsaicinoids, and codon-optimized T-DNA constructs for each of the ten cultivar specifications. I used Claude Code, cloud GPUs, and HPC clusters. Most of it happened at a desk, late, after a full day.
The work needed a home.
I filed the paperwork for Scoville Splice LLC because the science had gotten far enough that it deserved to exist properly — with real IP structure, real licensing, real data deposits. The ten cultivar specifications are complete: synthesis-ready DNA, academic papers, CRISPR guides, all deposited on Zenodo and Harvard Dataverse. The platform now extends the same pathway into 17 other plant species.
Scoville Splice runs alongside my work at Kronos Fusion — not competing with it, just occupying different hours. The greenhouse is still full of plants. The family still tastes everything.
Growing it — literally.
The constructs are designed. The next step is physical realization: Agrobacterium-mediated transformation, T₀ selection, homozygous lines, and HPLC validation of the SHU targets. That work will happen through licensed partners. In the meantime, the platform keeps extending — more organisms, more tiers, the cross-species framework now covering tomato, eggplant, potato, and beyond.