From Tokenmaxxing to Trusted Throughput — Mingsheng Hong, Ironclad
Mingsheng Hong from Ironclad warns against measuring AI usage as a competition—it only leads engineers to care about using more tokens rather than fewer.
Ironclad uses the same metrics, but as a smoke alarm: if a team uses surprisingly few tokens, it's worth discussing. The important thing is not just measuring cost, but also value—Ironclad now tracks "trusted throughput," meaning work that passes code review, automated tests, and reaches actual customers. The bottleneck often lies not in AI cost but in slow review processes and testing, so Hong advocates for unglamorous improvements like fixing unreliable tests and reducing agent retries.
Vibekollen prepared this summary with AI from the original publication. The content belongs to AI Engineer.
More to read
Lumaril, which develops AI software for optimizing industrial production by connecting to existing control systems, raised a $66M Series A at a $400M valuation
Techmeme 4 h ago
DeepSeek open sources Super-Agent as a research prototype, a proactive coding agent it says can monitor group chats for coding issues and suggest solutions
Techmeme 5 h ago
Insurance software maker Guidewire agrees to acquire Hyperexponential, which makes AI-powered insurance underwriting and pricing software, for $1.7B in cash
Techmeme 5 h ago
One word for AI agents in 2027
Nate Herk 9 h ago