이 에피소드에 관해
Blake Eastman tried rebuilding the SaaS products his company depended on—from Riverside and Asana to bookkeeping software—using AI.
The tools worked for his personal workflow. Then he encountered the difference between building something for yourself and operating a real software company: infrastructure, edge cases, observability, maintenance, design, guardrails, and thousands of users.
In this episode:
- Why Blake became less eager to replace mature SaaS products
- The hidden work required to turn an AI prototype into reliable software
- How he decides what to build and what to leave alone
- Why he now prefers products with APIs, CLIs, and MCPs
- How PostHog and Sentry expose errors, usage, and LLM costs
- The 85% rule he uses to avoid endless AI experimentation
- How builder critiques reveal what polished productivity advice misses
- Why Blake believes “everything is connectable”
- Inside his source-bound research system, the Academic Holocron
Blake’s conclusion isn’t to stop building. It’s to build narrow tools that create personal leverage, respect the complexity of mature infrastructure, and connect the best existing products into a system shaped around how you actually work.
Already Here explores how people are already living and working with personal AI—and what changes when these tools become part of everyday life.
Links mentioned in this episode:
The Nonverbal Group: https://www.nonverbalgroup.com/
Blake’s newsletter: https://www.nonverbalgroup.com/newsletter
Blake on X: https://x.com/blakeeastman
Michael on X: https://x.com/msg
Riverside: https://riverside.com/
Asana: https://asana.com/
Bench: https://www.bench.co/
PostHog: https://posthog.com/
Sentry: https://sentry.io/welcome/
Warp: https://www.warp.dev/terminal
Aside: https://aside.com/
Upwork: https://www.upwork.com/
Claude Code: https://claude.com/product/claude-code
OpenClaw: https://openclaw.ai/
Lettera: https://lettera.md/
Agentation: https://www.agentation.com/
