GenAI PMMarc Baselga
An AI/PM commentator quoted on internal AI workflows and measurement. The newsletter attributes to him the idea of companies overlooking their internal AI factory.
Key Highlights
- Marc Baselga is cited as a practical voice on internal AI workflows, adoption strategy, and how PM teams should evaluate AI impact.
- He argues that companies often miss their internal AI "factory" by tracking token usage and tool adoption instead of business outcomes.
- His commentary consistently supports giving PMs access to agentic coding tools like Claude Code and Cursor for prototyping and codebase understanding.
- He emphasizes that strong AI adoption depends on leadership behavior, low-friction infrastructure, and clear safety policies.
- His examples show that even non-technical product leaders can use modern AI tools to create high-leverage internal workflows.
Marc Baselga
Overview
Marc Baselga appears in the newsletter as a recurring commentator on how AI is changing product work, internal workflows, and enterprise adoption. His comments consistently focus less on AI hype and more on operating reality: how teams actually use tools like Claude Code, Cursor, copilots, and MCP-based systems; how leaders should measure success; and where organizations create friction by optimizing for visible participation instead of business outcomes.For AI Product Managers, Baselga matters because his advice sits at the intersection of product practice, internal tooling, and organizational behavior. Across the mentions, he frames AI not just as a feature layer but as an internal production system—an "AI factory"—made up of agent-driven workflows, code access, governance, and decision processes. His perspective is especially relevant for PMs responsible for AI adoption, developer workflow transformation, and measuring whether AI investments are producing meaningful outcomes.
Key Developments
- 2026-01-17: Marc Baselga shared best practices for Product Customer Advisory Boards, emphasizing that Product CABs should be distinct from Sales CABs and used to collect unfiltered customer input rather than present the roadmap.
- 2026-03-03: He noted that product leaders increasingly see lack of Claude Code access to repositories as a red flag when evaluating companies, because repo-connected AI tools can answer deep code questions without long engineer handoffs.
- 2026-03-13: He recommended quantifying the cost of slow AI adoption—such as missed markets, lost deals, and compliance delays—and finding a senior IT or C-suite sponsor to safely expand tool access beyond limited deployments like Copilot.
- 2026-03-22: He warned that many teams measure AI adoption using token counts, connector hits, or other activity metrics rather than tracking real business outcomes.
- 2026-03-26: He shared a reading list for product leaders, including Benedict Evans on competitive moats in AI and Gokul Rajaram on AI-native organizations potentially collapsing traditional product, design, and engineering boundaries.
- 2026-04-04: He argued PMs should have access to agentic coding tools such as Claude Code and Cursor for prototyping, codebase querying, and turning specs into working artifacts, while cautioning that direct production push access is a more complex governance question.
- 2026-04-08: He highlighted an Adobe product lead with no coding background who used Claude Code plus a folder of markdown files to create an AI chief of staff, illustrating how non-engineers can operationalize AI workflows.
- 2026-04-22: He warned that even as tools like Claude Code compress build cycles to hours, decision-making still often depends on status and confidence rather than evidence, increasing the risk of expensive late-stage mistakes and churn.
- 2026-05-11: He observed uneven AI adoption across 400+ senior product leaders in Supra, arguing that many companies optimize for participation over outcomes, while top performers combine visible leadership usage, low-friction infrastructure, and clear safety policies.
- 2026-05-13: He said companies often overlook their internal AI "factory"—the agent-driven workflows formed by Claude Code, Cursor, MCPs, and copilots—and instead focus on token usage or tool adoption metrics.
Relevance to AI PMs
1. Measure AI by outcomes, not activity. Baselga repeatedly warns against using token counts, connector hits, or adoption dashboards as proxies for success. AI PMs can apply this by tying internal AI programs to cycle-time reduction, win rates, quality improvements, support deflection, or revenue impact.2. Treat internal AI workflows as a product system. His "internal AI factory" framing is practical for PMs building org-wide AI capabilities. Instead of evaluating tools one by one, PMs should map the full workflow: repo access, context sources, MCP integrations, review loops, safety controls, and handoff points.
3. Expand access carefully but intentionally. Baselga’s comments suggest PMs should push for broad access to agentic tools for prototyping and code understanding, especially for non-engineering product work, while also defining governance boundaries around production changes, approvals, and safety policies.
Related
- Adobe: Referenced through an example of an Adobe product lead using Claude Code to build an AI chief of staff despite having no coding skills.
- Claude Code: Central to many Baselga mentions, especially around prototyping, repo access, non-engineer enablement, and internal workflow acceleration.
- Cursor: Mentioned alongside Claude Code as an agentic coding tool PMs should have access to.
- Agentic coding: A core theme in his commentary, especially the shift from static docs and specs to AI-generated working artifacts.
- Benedict Evans: Appears in Baselga’s curated reading list on AI strategy and moats.
- Gokul Rajaram: Also featured in Baselga’s reading list, particularly around AI-native org design and the possible collapse of traditional CPO structures.
- AI adoption: One of Baselga’s main themes, especially uneven rollout, executive sponsorship, and adoption measured against outcomes.
- Token counts: Used by Baselga as an example of misleading AI success metrics when divorced from business impact.
- Copilot: Mentioned as a baseline tool that some companies stop at, even when broader AI tooling may be needed.
- Product CABs / Sales CABs: Baselga distinguishes these to emphasize better customer signal collection and strategic product learning.
- Investor selection filters: His observation that candidates view lack of modern AI tooling as a red flag implies AI readiness is becoming part of company evaluation.
- Supra: Referenced in connection with his observation about uneven AI adoption across 400+ senior product leaders.
- MCPs: Included in his description of the internal AI "factory" made of connected agent workflows and infrastructure.
Newsletter Mentions (11)
“#14 in Marc Baselga says companies often ignore their internal AI “factory”—the agent-driven workflows from Claude Code, Cursor, MCPs and copilots—and instead measure token usage or tool adoption.”
#14 in Marc Baselga says companies often ignore their internal AI “factory”—the agent-driven workflows from Claude Code, Cursor, MCPs and copilots—and instead measure token usage or tool adoption.
“Marc Baselga observes that AI adoption is uneven among 400+ senior product leaders in Supra as companies optimize for participation over outcomes; top performers combine leadership AI usage with easy-to-use infrastructure and clear safety policies.”
#6 in Marc Baselga observes that AI adoption is uneven among 400+ senior product leaders in Supra as companies optimize for participation over outcomes; top performers combine leadership AI usage with easy-to-use infrastructure and clear safety policies.
“in Marc Baselga warns that although tools like Claude Code let teams prototype in an afternoon and ship to staging before lunch, decision-making still hinges on status and confidence, leading to costly, late-detected mistakes and higher churn.”
#17 in Marc Baselga warns that although tools like Claude Code let teams prototype in an afternoon and ship to staging before lunch, decision-making still hinges on status and confidence, leading to costly, late-detected mistakes and higher churn.
“Marc Baselga shows how an Adobe product lead with zero coding skills set up Claude Code to turn a folder of markdown files into an AI chief of staff.”
#21 in Marc Baselga shows how an Adobe product lead with zero coding skills set up Claude Code to turn a folder of markdown files into an AI chief of staff.
“#12 in Marc Baselga argues PMs should absolutely have agentic coding tools (e.g., Claude Code, Cursor) to prototype, query the codebase, and turn specs into working artifacts—yet granting them direct push access to production remains a far more complex debate.”
GenAI PM Daily April 04, 2026 GenAI PM Daily 🎧 Listen to this brief 3 min listen Today's top 17 insights for PM Builders, ranked by relevance from X, Blogs, and LinkedIn. Claude subscriptions will no longer cover usage on third-party tools like OpenClaw. #12 in Marc Baselga argues PMs should absolutely have agentic coding tools (e.g., Claude Code, Cursor) to prototype, query the codebase, and turn specs into working artifacts—yet granting them direct push access to production remains a far more complex debate.
“#17 in Marc Baselga shares 5 sharp reads for product leaders this month. Highlights include Benedict Evans’ case that OpenAI lacks a durable moat and Gokul Rajaram’s prediction that AI-native firms will eliminate the traditional CPO role by merging product, design, and engineering.”
#17 in Marc Baselga shares 5 sharp reads for product leaders this month. Highlights include Benedict Evans’ case that OpenAI lacks a durable moat and Gokul Rajaram’s prediction that AI-native firms will eliminate the traditional CPO role by merging product, design, and engineering. #18 in Dharmesh Shah echoes Reid Hoffman’s insight that AI-powered agents open vast new opportunities for software companies, proving software is far from dead.
“#10 in Marc Baselga warns that many product teams are gauging AI adoption through usage stats—token counts, connector hits or even weekly “demo spinner” games—rather than tracking real business outcomes.”
A product measurement insight emphasizes outcome-based AI adoption tracking. #10 in Marc Baselga warns that many product teams are gauging AI adoption through usage stats—token counts, connector hits or even weekly “demo spinner” games—rather than tracking real business outcomes.
“Marc Baselga recommends quantifying the cost of slow AI adoption (missed markets, lost deals, compliance delays) and enlisting a senior IT- or C-suite sponsor to push for safe approval of broader AI tools beyond just Copilot.”
#15 in Marc Baselga recommends quantifying the cost of slow AI adoption (missed markets, lost deals, compliance delays) and enlisting a senior IT- or C-suite sponsor to push for safe approval of broader AI tools beyond just Copilot.
“#21 in Marc Baselga notes product leaders now see lack of Claude Code access to repos as a red flag when choosing a company.”
#21 in Marc Baselga notes product leaders now see lack of Claude Code access to repos as a red flag when choosing a company. Connecting Claude Code lets PMs get instant, structured answers to deep code queries instead of lengthy engineer discussions. #22 in Greg Isenberg urges PMs to rebuild every SaaS tool—Notion, Slack, Stripe, etc.—as agent-native (payments, communication, memory) because the coming machine-to-machine economy will feature billions of software agents as customers.
“Running Effective Customer Advisory Boards: Marc Baselga shared best practices for Product CABs (Customer Advisory Boards). He advises distinguishing Product CABs from Sales CABs—Product CABs exist to gather unfiltered input, not to showcase your roadmap.”
From LinkedIn • Deeper Insights Product Management Insights & Strategies Framework for AI Agent Success: Dharmesh Shah introduced a clear formula—Agent Success = IQ × EQ × CQ—to evaluate AI agents. IQ measures reasoning ability, EQ measures collaboration skills, and CQ (Context Quotient) reflects how well the agent knows your business context. Dharmesh emphasizes that a model’s raw intelligence is useless without context and advises feeding agents with your top rep’s call recordings, NPS comments, renewal notes, and other internal signals to drive real impact. ( Dharmesh Shah ) Running Effective Customer Advisory Boards: Marc Baselga shared best practices for Product CABs (Customer Advisory Boards). He advises distinguishing Product CABs from Sales CABs—Product CABs exist to gather unfiltered input, not to showcase your roadmap. Key tips include requiring members to commit to at least one pilot per year, selecting forward-thinking customers (not just your largest accounts), and using CAB sessions to debate real strategic bets like pricing and new directions. ( Marc Baselga ) AI Tools & Applications Automating Project Memory: Peter Yang offered a simple yet powerful prompt: after a long chat with your AI agent, ask “Update project memory with the takeaways from this chat.” This approach automates the persistence of context—no more manual copy-and-paste—and keeps your projects aligned as they evolve. ( Peter Yang ) AI Industry Developments & News One-Click AWS Databases in Vercel for Agent-Driven Development: Guillermo Rauch announced that Vercel now enables one-click provisioning of AWS databases—RDS, DynamoDB, and Aurora DSQL—directly from the platform. This integration removes manual setup of environment variables, secrets, and network configuration, paving the way for seamless, production-grade agentic coding where AI handles backend infrastructure for you. ( Guillermo Rauch )
Related
Anthropic's coding assistant used for programming and automation tasks. The newsletter references it for building a custom approval device and for writing and research workflows inside AI agents.
An AI coding editor and automation platform. The newsletter highlights multi-repository support for automations across codebases.
An AI development pattern where models act more like autonomous coding agents. The newsletter uses it to describe both NVIDIA Dynamo’s target workload and GPT-5.5/Codex improvements.
Stay updated on Marc Baselga
Get curated AI PM insights delivered daily — covering this and 1,000+ other sources.
Subscribe Free