V-Sum Twenty
Tuesday, May 16, 2023
Technical Briefings
Shamir Kharkel
Shamir Karkal, co-founder of Sila and previously Simple and BBVA\'s API platforms, demos the Sila money API — a developer-first platform built around three primitives (users, wallets, transactions) that lets any builder onboard users, run KYC, link bank accounts, and move ACH. The demo app at demo.silamoney.com onboards both individuals and businesses via unique alphanumeric handles; KYC runs asynchronously through Socure/LexisNexis (3–15 seconds typical) with document-upload fallback for IDs or formation docs. Every user gets one wallet by default but can own many (one-to-many), enabling segregated-account patterns. Sila supports three ways to link bank accounts: raw routing+account entry, your own Plaid integration via processor tokens, or Sila\'s managed Plaid instance (Plaid sandbox user_good/pass_good). Same-day ACH pulls $1,234 into the wallet in roughly five minutes in sandbox — deliberately slow so developers don\'t build code expecting real-time settlement. Wallet-to-wallet transfers settle on Sila\'s ledger in seconds; external redeems push ACH out to any linked bank account.
NVAL demos pricing infrastructure for illiquid assets, starting with NFTs. Using AI and on-chain data feeds, NVAL produces fair market values for NFTs in near real time — foundational for any financial workflow (lending, borrowing, insuring, trading) on non-fungible assets that have no spot price. NVAL sits as an analytics layer between blockchains (L1s, L2s, cross-chain) and the industries building on top (gaming, sports, trade finance, accounting, lending). Mark (CTO) walks through the REST API documented with OpenAPI: a "try it out" docs UI, a Postman collection, and a CLI all return the same price prediction for a specific NFT (the demo values a Doodle at 3.3 ETH with 96% confidence, sitting between the listed price of 4.2 ETH and the highest offer of 2.8 ETH). A Chrome extension surfaces NVAL prices on OpenSea pages, and the NVAL web app shows floor price, listed price, and price history against actual trade prints. The price-history endpoint returns 52 weekly data points across a year (configurable daily). Feature importance breaks down what drives price — 96% market action for a Doodle, 72% market action + 30% accessory-trait for a CryptoPunk — giving traders a way to understand why NVAL priced an NFT where it did.
Amelia Niemi, co-founder and CTO of Ntropy, walks through the Ntropy transaction enrichment and categorization API. Raw bank transactions go in, structured data comes out — category, merchant identity, logo, location, website, recurrence pattern, active and cancelled subscriptions, and detected income sources. A live Python quickstart shows four lines of code enriching a 1,500-row CSV (roughly 12–18 months of transaction history); the Ntropy app then surfaces the same data in a PFM-style UI. Amelia walks through Ntropy's benchmarks against GPT-4 — about 2,000× cheaper per transaction, 150× faster, and 15× more accurate on the hardest cases — before ML engineer Robin demos "Ntropy Cookie," a Discord chatbot that lets end users ask natural-language questions about their own spending: essentials vs non-essentials, savings opportunities, bank fees.