Jacob Silverman

AI Engineer — agentic systems over financial data

Summary

Ten years building data-intensive systems, the last four on AI: local multi-modal pipelines, agentic harnesses with evals and guardrails, and the deterministic engines they reason over. Full-stack throughout — the pipelines, the engines, and the surfaces people use.

Experience

Software Engineer · Self-employed

2020Present

Fairway — Claude Code’s docs tell you what to do when you ask; Fairway tells you what to do when you don’t know what to ask, and gives you hooks to encourage patterns and discourage antipatterns. Nine small devtools sharing one Go binary, one WAL-mode SQLite store and one MCP server, so anything one tool writes another can read — each extending a command Claude Code already ships: /insights, /usage, claude agents and the tool hooks. Pick the ones you want: most are read-only pull commands that need nothing installed, and two wire into your setup with explicit install and uninstall verbs. The binary alone creates no files, opens no ports, writes no config. https://github.com/bespy/fairway

Ledger AI — high-volume financial reporting for crypto and fiat. Automated P&L, balance sheet and cashflow statement, with an upsell to a personalised investigation service. Crypto users connect wallets and Fiat users connect Xero, MYOB or MEWS (hotels). High volume, hundreds of thousands of on-chain multi-chain events parsed, tens of thousands of fiat transactions reconciled for clients to date. https://ledgerai.fly.dev

Ledger UI — a React component library of animated accounting primitives with spreadsheet input semantics: journal entries, reconciliation matching, P&L and balance-sheet tables. https://ledgerui.fly.dev

Shared engine — a deterministic double-entry engine that books crypto and fiat from raw events with full provenance; a multi-modal ingestion pipeline running local models over voice, vision and OCR; and an agentic harness where the model proposes and the engine computes, with eval datasets grading every proposal. Built over six years of trading infrastructure on Solana — P&L analytics and trade-ledger reconstruction for delta-neutral trading and DLMM market making (Meteora).

Freelance — work including in regulated industries: an approval engine for a regulated-sector client, with legally binding deadlines, scoped directly with the non-engineers who own the process.

Sales Engineer · STRV

20172018

Pre-sales engineering at a US software-development agency (New York) — scoping custom software builds and translating client requirements into project specs.

Software Engineer Intern · Atlassian

20142015

Software engineering internship in the Jira/Confluence ecosystem, during my UNSW engineering degree.

Education

University of New South Wales

20122016

Bachelor's Degree in Engineering (Electrical)

Tech Stack

Agentic AI / LLM tools · RAG / semantic search · Tool calling / function calling · Prompt & context engineering · Evals / structured output · Local LLMs (Llama, MLX) · Multi-modal ML (Whisper, Qwen3-VL, OCR) · Inference optimization (quantization, batching) · MCP · TypeScript · Python · Rust · React · Node.js · Remix / React Router · Next.js · Data viz / charting (D3.js, SVG) · Tailwind · Radix / shadcn · Zod · PostgreSQL · Redis · SQLite / FTS5 · ORMs (Prisma, Drizzle) · Solana (Anchor, Helius) · On-chain forensics · Xero & Stripe APIs · Inngest (durable workflows) · Docker · Kubernetes / Helm · CI/CD (GitHub Actions) · Testing (Vitest, Playwright) · Sentry · GCP · AWS · Fly.io