Monzo engineers demo a trustworthy financial agent: 5 design principles for AI you can rely on
AI Engineer · youtube · 2026-10-17
In a 110-minute AI Engineer workshop, Monzo engineers Francesco Galletta and Felippe Felisola Caso live-build a financial-analyst agent harness around one real problem: budgeting a wedding. They present five design principles for trustworthy AI — shared domain, consistency, evidence, accountability and verification — arguing math belongs in code, long linear chats fall apart, and verification needs golden examples and external reference points. Q&A covers how many harnesses to build, when to stop over-sharpening the axe, and skills vs. dedicated harnesses.
More from coding & agent
- Open-source Claude Code skill embeds a live website inside a tweet via player cards — doooyle · 2026-10-17
- YC Partner: Personal Software Is Cheaper Than Ever, and Open-Sourced Code Lets Others' Agents Fill the Gaps — ycombinator · 2026-10-17
- Claude Code's contextual UX hints earn praise from a heavy daily user — iamsahaj_xyz · 2026-10-17
- Codex-created worktrees blamed for killing a 600GB-battered SSD; users say disk bloat is the worst part of AI coding tools — natesiggard · 2026-10-17
- Engineer: I'm not smarter than AI anymore — time to learn to orchestrate it — KevinNaughtonJr · 2026-10-17
- Open-source OODA pentest agent drives Kali and Playwright tools directly over MCP — irraz_rulez · 2026-10-17