
Velocity Through Rigour: Then & Now • Cate Huston • Bug Bash EU 2026
Title history
- Velocity Through Rigour: Then & Now • Cate Huston • Bug Bash EU 2026 — observed on 20261028-0850 UTC (YouTube)
Keywords
Summary
153 words
Critical Evaluation
Value of the Information & Strength of the Argument
The talk provides valuable, hands-on insights from a seasoned engineering leader who has personally navigated the shift to AI-driven development. Huston’s argumentation is solid: she builds a clear case that the fundamental principles of engineering management—clarity of purpose, well-defined processes, and effective guardrails—are more relevant than ever, even when the ’team’ consists of humans and AI agents. She supports her views with concrete examples from her recent re-platforming project, including the evolution of guardrails and the rationale behind specific choices like release branches and merge queues. The 30% guardrails heuristic is an interesting and actionable takeaway, though it is presented as an intuition rather than a rigorous metric. The argument that managers are force multipliers and that token-spending rhetoric needs to be redirected toward business value is persuasive and timely.
Scientific Rigor, Source Quality, Title Accuracy
The talk is based on the speaker’s direct experience, which lends authenticity but limits generalizability. She references external data from Meta’s randomized trial and DORA 2025, but does not provide specific metrics or citations, leaving the audience to trust her interpretation. The title is well-aligned with the content: the ’then vs now’ comparison is explicit and the emphasis on ‘velocity through rigour’ is consistently demonstrated through her examples. No commercial sponsorship or advertising is present; the only promotional element is a brief mention of her own book and career platform, which is relevant to the topic. The overall rigor is adequate for a conference talk, but viewers seeking hard evidence would need to consult the original studies mentioned.
263 words
Title / Content Match
The title accurately reflects the talk's theme: it contrasts prior engineering practices ('then') with current AI-driven approaches ('now') while emphasizing how rigour, in the form of guardrails and process, enables velocity.
Quality & Reliability
7/10
The talk is grounded in the speaker's extensive practical experience but relies on anecdotal evidence and a few referenced studies (Meta, DORA) without deep detail. The insights are coherent and credible, though not fully verifiable from the content alone.
Key Moments
Markers derived by PSI from the transcript: the creator did not define chapters.
- Huston introduces two key questions: percentage of output should be guardrails, and what of traditional working methods remain relevant amid industry upheaval.
- Describes the re-platforming arc from April to September, highlighting the transition from Bubble to a proper database and the importance of preparatory work.
- Discusses the metaphor of pampered raccoons and challenges the apparent devaluation of management, arguing that managers are essential force multipliers.
- Outlines the four layers of management (mission, strategy, tactics/process, execution) and how they have accelerated in the AI era, emphasizing process as key leverage.
- Explains the guardrail concept and the 30% heuristic, sharing a chart showing guardrail evolution and the need for instrumentation and automation.
- Concludes with architecture matters earlier than ever, highlighting design systems, role-based access control, and the amortization of reliability work.
Cited Sources
- Bug Bash EU 2026 — Official conference page where the talk was presented.
- The Engineering Leader — Author's book referenced as a recommended resource for leadership.
- DIY Your Career — Platform co-built and maintained with AI assistant Claude, used as a case study.
- Cate Huston on LinkedIn — Speaker's professional profile for reference.
External References
Contribution & Novelties
The talk offers a practical framework for applying engineering-management rigour to AI-assisted development, including a 30% guardrail heuristic and a layered model of mission, strategy, tactics, and execution. It bridges traditional team leadership with agent-heavy workflows, showing how small teams can adopt scaled-up practices, and emphasizes that guardrails should be automated and instrumented rather than merely written down.
Pour aller plus loin :
- DORA — DevOps Research and Assessment reports on software delivery performance, referenced for throughput/stability trade-offs.
- Shift-left testing — Moving testing earlier in the development cycle, a key enabler for guardrails like the one described.
- Engineering management — The discipline of managing technical teams, here adapted for AI-era velocity.
111 words
Radar Profile
The radar shows a strong profile with high information quantity (8) and moderate quality/reliability (7 each), indicating a dense but presentation-heavy talk. The balance between quantity and technical level suggests a talk that is comprehensive yet accessible, with reliability trailing slightly due to anecdotal evidence.