-
Prompt Like a Pro: Three Battle-Tested Tips for Amplitude Global Agent Success
Read more: Prompt Like a Pro: Three Battle-Tested Tips for Amplitude Global Agent SuccessGreat prompts turn Amplitude Global Agent into a dependable analyst you can trust. By defining the role, clarifying the objective, and setting guardrails, you reduce ambiguity and improve output quality. Grounding the agent with concrete data context and examples sharply lowers hallucinations and speeds up analysis. Iterating with A/B testing and lightweight evals replaces guesswork…
-
Beyond Accuracy: How I Evaluate AI Customer Service Agents That Delight and Scale
Read more: Beyond Accuracy: How I Evaluate AI Customer Service Agents That Delight and ScaleMost teams judge AI agents by accuracy and resolution rates—but that’s only half the story. In my POCs, I mirror real-world messiness, stress-test multi-turn and multilingual scenarios, and watch how the Agent recovers, stays on track, and hands off to humans. I also evaluate the conversation design and brand tone because customers don’t experience “accuracy,”…
-
From Ed‑Tech Roots to Core Analytics: Product Leadership Lessons Inspired by Amplitude
Read more: From Ed‑Tech Roots to Core Analytics: Product Leadership Lessons Inspired by AmplitudeGreat analytics products are built by PMs who pair startup scrappiness with rigorous discovery and outcome-based execution. Drawing inspiration from Amplitude’s core analytics product, I share how driver trees, continuous discovery, and unified event taxonomies translate into faster learning cycles. The post outlines practical tactics for activation, retention analysis, and A/B testing, all anchored in…
-
Stop Losing Users: How a Second Message and Prompt Audit Drive 2–3x Retention
Read more: Stop Losing Users: How a Second Message and Prompt Audit Drive 2–3x RetentionDefault prompts are silently depressing your agent’s performance by creating ambiguity and friction after the first reply. I’ve found that adding a fast, well-designed second message—paired with a structured audit of every default prompt—consistently drives a 2–3x retention lift. The key is to clarify scope, offer concrete choices, and tune for channel (voice vs. chat)…
-
Old-School Selling Beats PLG in the AI Era: My GTM Playbook for 8‑Day Enterprise Deals
Read more: Old-School Selling Beats PLG in the AI Era: My GTM Playbook for 8‑Day Enterprise DealsOld-school, in-person selling still beats product-led motions in the AI era because enterprise buyers reward teams who show up and de-risk change. I share how to sell to engineers and executives in one motion, compress cycles for AI-native buyers, and navigate build vs. buy with speed and clarity. You’ll learn why process should raise the…
-
Supercharge Core Web Vitals with Amplitude’s Global Agent: Faster Rankings, Happier Users
Read more: Supercharge Core Web Vitals with Amplitude’s Global Agent: Faster Rankings, Happier UsersCore Web Vitals are a direct lever on user experience and SEO, so I use Amplitude’s Global Agent and Amplitude AI Agents to measure and improve them in near real time. In this post, I share a pragmatic playbook that unifies instrumentation, segmentation, and experimentation to translate performance gains into rankings and revenue. You’ll see…
-
My Always‑On AI Team: How I Get Claude Agents to Tackle Work While I’m Offline
Read more: My Always‑On AI Team: How I Get Claude Agents to Tackle Work While I’m OfflineI run an always-on team of Claude-powered agents that prep my podcasts, streamline sales follow-ups, and deliver weekly coding retrospectives—so work moves forward even when I’m offline. The system is inspired by OpenClaw but designed with stricter guardrails to avoid surprise bills and unsafe access. Four components make it reliable: explicit agent identities, a MacOS…
-
Inside My Pricing Playbook: Building Value-Based Packaging That Balances Growth and Profit
Read more: Inside My Pricing Playbook: Building Value-Based Packaging That Balances Growth and ProfitPricing and packaging is never just a price tag—it’s a strategic system that aligns customer value, market reality, and business outcomes. In this playbook, I share how I move from foundational qualitative research to quantitative willingness-to-pay studies, then translate those insights into robust financial models. You’ll see how I use methods like Gabor-Granger and Van…
-
Built for Your Biggest Days: How We Engineer Fair, Reliable Scale Without Downtime
Read more: Built for Your Biggest Days: How We Engineer Fair, Reliable Scale Without DowntimeEnterprise teams ask sharper questions about scale—and they should. I share how we handle 150k+ requests/sec, shard our source-of-truth data with Vitess and PlanetScale, and reshape Elasticsearch online without downtime. You’ll see how we isolate spikes in a multi-tenant architecture, route AI workloads across LLM providers, and use heartbeat metrics, CI/CD, and feature flags to…
-
From PM to AI Engineer: RAG, Evals, and Discovery—The Surprising Playbook I’m Applying
Read more: From PM to AI Engineer: RAG, Evals, and Discovery—The Surprising Playbook I’m ApplyingThis episode is a masterclass in how modern product managers can step confidently into AI engineering without a traditional software background. You’ll hear a pragmatic breakdown of context engineering, prompt writing, RAG, observability, and evals—and why discovery skills transfer directly. I share how I’m applying a retrieval-first pipeline, eval-driven development, and vibe coding discipline with…
-
Stop Asking AI Anything: The 3 Outcome-Based Prompts That Unlock Real Product Insights
Read more: Stop Asking AI Anything: The 3 Outcome-Based Prompts That Unlock Real Product InsightsGeneric AMAs produce generic answers. Instead, I frame every AI interaction around three outcome-based prompts: the business result and metric I’m targeting, the exact context and constraints to analyze, and the decision plus deliverable I need to act. This approach turns behavioral analytics into precise recommendations tied to outcomes vs output OKRs. It also speeds…
-
Level Up: May 26 Claude Code Show & Tell + Final Product Discovery Fundamentals Cohort
Read more: Level Up: May 26 Claude Code Show & Tell + Final Product Discovery Fundamentals CohortTwo timely opportunities to strengthen your product practice are on the calendar. On May 26 at 9am PDT, join a lively Claude Code: Show & Tell to swap proven AI workflows and leave with practical playbooks. Invites will go to Supporting Members and CDH Members—watch your inbox. Then, from June 4 to July 9, I’m…
Signup for Weekly Digest Emails
Categories
- AI Strategy (294)
- Generative AI (190)
- Generative AI, Product Management (1)
- IT Leadership (74)
- Leadership (166)
- Product Management (394)
- Product Management Leadership (282)
- Uncategorized (201)
Archieve
- June 2026 (15)
- May 2026 (63)
- April 2026 (51)
- March 2026 (44)
- February 2026 (41)
- January 2026 (49)
- December 2025 (79)
- November 2025 (64)
- October 2025 (263)