Build betterproducts.Lead what's next.
Get closer to customers. Turn ideas into working products. Put AI to work across your team. These are the methods I use, shared for product builders and the leaders behind them.
Free handbook, agent blueprints, and templates. No signup.
Built in product teams. Shared for yours.
I'm Falk Gottlob. Thirty years building products, four CPO roles, and five exits. Now building Falkster.ai.
- Microsoft
- Adobe
- Salesforce
- Commure
- Crisis Text Line
- SOCi
- Smartcat
Learn the method. Put it to work.
Three ways to move your product work forward. Start with the method, the agent, or the artifact you need.
Make better product decisions.
From customer discovery to shipping and leadership. A working playbook drawn from how I build and run product teams.
Discovery · Prototyping · AI craft · Leadership
Give your team more capacity.
Turn recurring product work into reliable agent workflows. Each blueprint includes a prompt, an eval, and a shipping gate.
Sense · Discover · Decide · Build · Ship · Measure · Amplify
Move from thinking to doing.
Start with a useful artifact. Adapt it to your team and put it to work.
Prompts · Evals · Discovery scaffolds · Team rituals
Go deeper.
Take the playbook with you.
Print it, send it to your team, read it on a plane, or load it into Claude.
- 01The Thesis
- 02Discover
- 03Build & Ship
- 04AI Craft
- 05Lead the Shift
- 06Forward Deployed
The Design Operating Model
Connect customer needs, product behavior, and design decisions.
The FDE Transition
The engineer who owns the customer's outcome, and what product has to change first.
What's your Builder level?
Find the skill to work on next, with a clear starting point.
Product Builder in 7 days
One skill a day, one task a day.
Put the notebook to work with your AI.
The Falkster Corpus is the whole site as markdown. Drop it into a Claude Project, a ChatGPT project, or Cursor, and work normally.
What I'm building. What I'm learning.
The practice keeps evolving. Here are the decisions, experiments, and lessons behind it.
The Pilot Was Approved on the Sixty
A course lesson prices a 120-person claims team at $60 a month, then shows three sensible decisions making it $2,106. Put four cost lines on the handshake.
The Career Out of the Seat: Draw the Exits, Not Just the Ladder
The FDE career ladder is unsettled, and founders are now walking into the seat for paid work. Three exits to write into the offer, and one rule for the ladder.
Find Your Strikethrough: The Agent Shortlist Panel
Jason Lemkin's agent picked Render and struck Vercel off in one phrase. No lost deal in any CRM. A biweekly panel of workload questions finds your phrase.
Agents Fit the Human Pattern Nine Times in Ten. Design the Tenth.
Atlassian's rule: agents follow the patterns built for humans, nine times in ten. I put an agent in an approver's seat. The tenth is the unwritten judgment.
Snyk Ran Its Agent Inside for a Year. Copy the Gate.
Snyk moved an internal support agent to every paying customer. The surface was not the only change. Copy the four-part gate, not the year inside.
Five Sessions, Then the Prompting Habit Sets
Users settle on 2 to 4 prompt templates within about 5 sessions. My onboarding agent lowered activation. Put a person in the first week instead.
Bring the practice to your team.
I work selectively with product leaders on org design, AI agents, and how their teams build.
Questions about this notebook
Who is Falk Gottlob?
Falk Gottlob is a product executive with 30 years shipping product at Microsoft Research, Adobe, Salesforce (Marketing Cloud / Quip / Slack), and four startups. He has had five exits including one IPO and one acquisition by Microsoft (MVC, 2009), and the largest at $6.5B with Commure (2018). He has been CPO at Commure, Crisis Text Line, SOCi, and Smartcat, and is the founder of Falkster.ai (agentic AI, built and run by AI agents, in stealth). He writes falkster.com, the working notebook on AI-era product management.
What is The PM Standard for the AI era?
Product Builder OS is Falk Gottlob's six-skill definition of the modern Product Builder role, priority-ordered: Rapid Prototyping, Customer Proximity, AI Fluency, Outcomes Thinking, Storytelling & Distribution, and End-to-End Ownership. It replaces the old Product Manager job description (write specs, run rituals, push handoffs to engineering) with a Product Builder role that ships working prototypes, writes evals before code, owns the launch narrative, and treats AI agents as teammates. The full standard lives at falkster.com/pm-standard alongside a 49-chapter handbook on how to live it day to day.
What is The PM Handbook?
The PM Handbook is a 49-chapter operating system for product managers in the AI era, organized around the six Product Builder OS skills: Rapid Prototyping, Customer Proximity, AI Fluency, Outcomes Thinking, Storytelling & Distribution, and End-to-End Ownership. Each chapter is a practice Falk Gottlob is actively running, not a retrospective written years after the fact. Free, no signup. Lives at falkster.com/handbook.
How does AI change the product manager role?
AI changes PM along five axes: (1) specs are dead, prototypes ship; (2) evals replace requirements docs as the unit of work; (3) every PM should ship code, not just write requirements; (4) agents become teammates, not features to build; (5) cycle time collapses from weeks to days. Product builders combine customer judgment with hands-on prototyping and clear measures of quality. Falkster's Product Builder OS documents the six skills that define the new role.
What is a Product Builder?
A Product Builder is the AI-era successor to the Product Manager. Where a PM wrote specs and ran rituals, a Product Builder ships working prototypes (often with Claude Code or similar), writes evals before code, owns one or more product surfaces end-to-end, can defend cost-per-request to finance, and treats AI agents as teammates with prompts, evals, and ship-readiness gates. Falkster.com defines the role across six skills (Product Builder OS) and a 49-chapter handbook on how to live it.
What are AI agent blueprints?
AI agent blueprints are productized recipes for AI agents that handle PM work: monitoring product health, summarizing customer signal, drafting weekly digests, triaging the inbox, prepping board decks, etc. Falkster ships 44 free blueprints across seven PM operating-system stages (Sense, Discover, Decide, Build, Ship, Measure, Amplify), each with a prompt, an eval rubric, and a ship-readiness gate. Live at falkster.com/handbook/ai-agent-army.
Is the falkster content free?
Yes. Every chapter, agent blueprint, and toolkit template on falkster.com is free to read and use, no signup required. There is an optional once-a-month newsletter (The Dispatch) for new chapters and field notes, but nothing is gated. Commercial redistribution of full articles is not permitted without permission.
Can I work with Falk Gottlob directly?
Selectively. Falk takes on a small, deliberate set of advisory engagements each year while running Falkster.ai full-time. Three formats: Product Org Audit (2 weeks, written report + 90-day plan), Eval & Agent Workshop (3 days on-site, build the harness with the team), and Board & Exec Reviews (quarterly with a written memo). One slot typically opens per quarter. By application at falkster.com/work-with-me.