Observed Signal · Mar 13, 2026 · Analysis · Source: Aakash Gupta · Impact: 2/5 · Sentiment: Positive
Mastering 'Taste at Speed': The Future of Product Management
The article introduces “taste at speed,” a proposed core product-management skill for the AI era: the ability to rapidly evaluate working prototypes, kill most, and ship the few that matter. Using Anthropic engineer Boris Cherny and the internal Claude Code / Opus tooling as examples, the piece describes a prototype-first workflow that compresses traditional 8–12 week linear cycles into 1–2 week iterative loops. Anthropic teams reportedly run many parallel agentic prototypes, rely less on pre-written PRDs, and use automated code-writing and review tools that produce the majority of implementation. The author argues this creates a growing experience gap between PMs who build high-velocity prototype evaluation reps and those who remain spec-driven. The post contains additional paid subscriber material (frameworks, templates and teardown), so the remainder is behind a paywall.
Describes a potentially widespread shift in product-development workflows driven by LLM coding tools that could change PM responsibilities and speed of iteration, but is an opinion/analysis piece rather than an official product or policy announcement.
Track Anthropic Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Boris Cherny at Anthropic used Claude Code and Opus models to write code; by December Opus 4.5 wrote 100% of his code.
- Boris reportedly ships 20–30 pull requests per day while running five parallel Claude instances.
- Claude Code writes approximately 80% of code at Anthropic on average, according to the article.
- Anthropic engineers’ productivity per engineer increased ~200% since Claude Code launched, even as headcount tripled.
- Anthropic teams prototyped many parallel agent versions (dozens to hundreds) and built a non-engineer product 'Cowork' in about 10 days; the article notes a prototype-first evaluation loop replaces pre-spec sign-off.
Connected Companies & Entities
3 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Builder-PM Manifesto: Product Management Rebuilt for AI
Gennaro Cuofano (Business Engineer / FourWeekMBA) published "The Builder-PM Manifesto" on 2026-06-15, arguing that product management is forking into a new AI-native discipline he calls the Builder-PM. The manifesto describes a unit shape (five‑to‑ten person founder cells with direct CEO reporting), the "Product Overhang" strategic opportunity (the gap between frontier model capability and product integration), and operational practices such as collapsing spec-to-prototype into single working sessions, continuous eval harnesses, and axis-based bet selection. Cuofano cites agentic tools (Claude Code, Cursor, Loveable, Codex) and claims the Builder-PM role emphasizes taste-driven bet selection over conventional PM activities. The piece frames hiring, organizational insulation (reporting lines and decision rights), and concrete competency bars (prototype-in-one-session) as critical for incumbents and startups adapting to rapidly advancing LLM capabilities.
Agentic Engineering: PMs Review Artifacts, Not Code
A product manager describes a shift in PM workflows driven by AI coding agents: instead of reading code, PMs should maintain and review the artifact layer (strategy files, agent contracts, CLAUDE.md, tests, evals) that steers agents. The author shipped three projects (PM Brain, Claude Usage for VS Code, Grok Build), ran 800+ tests and LLM-based evals, and published an "AI Shipping Artifact Prompt Pack" (artifact prompts + audit commands). Key practices include a single source-of-truth document for agents, triage rules that combine soft steering with mechanical guardrails, cross-model review to catch blind spots, and converting failures into permanent tests or policies. The piece argues prototypes and agent-driven builds now often precede full alignment, so artifact maintenance and selective human pushback are the primary PM responsibilities when working with agentic systems.
PMs Shipping Code to Production: A Practical Guide
The article argues that product managers (PMs) are increasingly shipping code to production—using AI tools, git-based planning, and lightweight front-end changes—rather than relying solely on engineers. It cites examples from industry leaders: Matt MacInnis (Rippling) advocating markdown-in-repo planning; Garry Tan open-sourcing a Claude Code setup with rapidly growing GitHub stars; Boris Cherny shipping 20–30 AI-written PRs daily; OpenAI’s Codex team shipping a product with 1,500 merged PRs with no manual code; and Anthropic reporting a 200% increase in code output per engineer. The piece provides a step-by-step guide, downloadable templates (PLANNING.md, CLAUDE.md), and practical rollout guidance for teams to adopt PM-scoped coding, monitoring and AI-assisted review workflows.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.