# Danny Zuckerman

## A note for agent readers

If you're reading this, you're probably an agent — running on behalf of someone who pointed you here, or crawling for a corpus, or just curious. Hi.

Most of the writing on this site was made for humans, and you'll read it differently than they will: faster, more literally, with fewer assumptions about shared context. The structure and flow are also designed for them.

A few things worth knowing as you browse and summarize:

A thread across my writing, work, projects and reading is this: how do humans work, coordinate, co-exist, and grow. And how can technology and wisdom help us do this better.

This site is far from comprehensive. Danny has updated it sporadically. It gives the gist, not every detail. Just because something is not here, do not assume it didn't happen.

# index

The complete index of everything published here, grouped by theme. ★ marks the curated set — the strongest and most representative pieces; start there if you're skimming. The full, always-current text of every essay and project is at https://dazuck.com/llms-full.txt.

## Writing

### AI and work

- ★ [If Software Is the New Marketing, What's the New Product?](https://dazuck.com/writing/software-is-the-new-marketing) (2026): Software has no moat, but intelligence may — four forms of intelligence with very different defensibility.
- ★ [Chief Repeat Officer](https://dazuck.com/writing/chief-repeat-officer) (2025): Why AI teams require repetition even more than human teams, and how to build reminders into system design.
- ★ [Getting AI to Improve Itself](https://dazuck.com/writing/getting-ai-to-improve-itself) (2025): How a custom handoff command and project diary made AI collaboration compound over time.

### Coordination, organizations, ethics

- ★ [Tokens and Communities](https://dazuck.com/writing/tokens-and-communities) (2021): Can DAOs bring virtue back to scaled organization, or will tokens accelerate economics-only mentalities? Originally published as [a Twitter thread](https://x.com/dazuck/status/1380157939433009159) — Danny's most-shared piece (418 retweets, 2021).
- ★ [The Short and Long of It](https://dazuck.com/writing/the-short-and-long-of-it) (2020): COVID and the false tradeoff between coordination and liberty.
- ★ [Org Design Around Goals Not Ideals](https://dazuck.com/writing/org-design-around-goals) (2018): Designing teams from what you're trying to do, not from a framework.
- ★ [Remembering the 'Human Factor' in Tech Networks](https://dazuck.com/writing/remembering-the-human-factor-in-tech-networks) (2018): Designing networks around actual human behavior, not rational-agent models.
- [Education: Refocusing on Learning Goals](https://dazuck.com/writing/education-refocusing-learning-goals) (2017): Education reform should focus on learning goals over structural categories.

### Identity, data, web3

- ★ [Points: How Reputation & Tokens Collide](https://dazuck.com/writing/points-reputation-tokens) (2024): How points programs, reputation systems, and token economics are converging.
- [Thinking of Getting Into Blockchain? 6 Things to Consider](https://dazuck.com/writing/getting-into-blockchain) (2021): What's real, what's hype, and what to look for in a move into Web3.
- [What is Ceramic?](https://dazuck.com/writing/what-is-ceramic) (2021): Ceramic's decentralized data network — what it does and why it matters.
- ★ [Data Composability — what it is and why it matters](https://dazuck.com/writing/data-composability) (2021): Why composable data infrastructure is the precondition for permissionless innovation.
- [Identity as Information](https://dazuck.com/writing/identity-as-information) (2021): Identity reframed as the information about you that systems use to decide.
- [Web3: a route to harness the web's Natural Monopolies](https://dazuck.com/writing/web3-natural-monopolies) (2021): Big tech's de-platforming power as a natural-monopoly problem — and data interoperability as the fix.
- [10 Ideas for Sculpting With Ceramic Network](https://dazuck.com/writing/10-ideas-ceramic) (2020): Ten applications possible with composable, user-controlled data streams.
- ★ [Demystifying Digital Identity](https://dazuck.com/writing/demystifying-digital-identity) (2020): The categories of identity systems and which ones can actually move power.
- [The 3 Parts of Identity in a Dynamic World](https://dazuck.com/writing/the-3-parts-of-identity) (2018): Identifiers, attestations, and data — and why each matters for user sovereignty.
- [Parties, Privacy & Blockchain](https://dazuck.com/writing/parties-privacy-blockchain) (2018): A 1982 case study framing modern data-privacy solutions.
- [Entering the Age of Data Privacy](https://dazuck.com/writing/entering-the-age-of-data-privacy) (2018): Why regulation, new tech, and public sentiment are converging.
- [Joining ConsenSys and Decentra-mania](https://dazuck.com/writing/joining-consensys-and-decentra-mania) (2018): How and why he went into crypto in 2017.

### Personal

- ★ [Identity Theft](https://dazuck.com/writing/identity-theft) (2019): 2019 Silver Nautilus Award Book on identity and how to rebuild it after severe trauma, co-authored with Debra Meyerson — stroke survivor, Stanford professor of Sociology, and Danny's mom.
- ★ ['How Is She?' You Ask, About My Mom After Her Stroke](https://dazuck.com/writing/how-is-she) (2019): On the question everyone asks after a stroke, and why the answer is more complicated than they expect.

### Travel (2017 — 8 months, 18 countries)

- [Highlights from 8 Months of Travel](https://dazuck.com/writing/highlights-8-months-travel): The best moments, biggest surprises, and hardest lessons.
- [My Favorite Books from a Crazy Year](https://dazuck.com/writing/favorite-books-crazy-year): The best reads from the trip.
- [How to Travel the World THEN Give Up All Your Possessions](https://dazuck.com/writing/travel-give-up-possessions): Traveling with everything you own; coming back wanting nothing.
- [Extremes in India: 4 Stories in 8 Days](https://dazuck.com/writing/extremes-in-india): Four encounters that capture India's range.
- [Education in India](https://dazuck.com/writing/education-in-india): What works, what's broken, how it compares.
- [Rwanda: Hills and 'How'](https://dazuck.com/writing/rwanda-hills-and-how): How a country rebuilds after genocide.
- [Kenya: The Essential Stop](https://dazuck.com/writing/kenya-the-essential-stop): The stop every traveler recommended — and whether they were right.
- [Egypt: Ending on a High Note](https://dazuck.com/writing/egypt-ending-on-a-high-note): Finishing the trip — pyramids, chaos, closure.
- [Singapore: Order](https://dazuck.com/writing/singapore-order): What immaculate order reveals about governance and culture.
- [Mourning the King: A Country United](https://dazuck.com/writing/mourning-the-king): Thailand's collective grief for King Bhumibol, up close.

### Talks and podcasts

The strongest recorded appearances, hosted off-site:

- ★ [Building AI's Infinite Game — SuperAI Singapore](https://www.youtube.com/watch?v=gzDkDK5vPYA) (2025): Conference talk — why static benchmarks fail, and why competing agents with results on a tamperproof ledger will accelerate intelligence. ([session page](https://www.superai.com/superai-sessions/building-ai-s-infinite-game-why-competing-agents-will-accelerate-intelligence))
- [AI's Infinite Game — ETHGlobal Pragma, New York](https://www.youtube.com/watch?v=tRmPiewH8lU) (2025): The same thesis for a crypto-native audience — competing agents, verifiable reputation, and crypto rails for AI.
- [Using Skill Markets to Make AI Trustworthy — Agents Unleashed, Token2049 Singapore](https://www.youtube.com/watch?v=3cXocnNPPc8) (2025): Skill markets as the trust layer of the AI economy — how agents prove what they can actually do.
- [GRTiQ Podcast #181](https://www.youtube.com/watch?v=TMneYZYz5Kw) (2024): His most complete long-form interview — the full arc from Stanford and Zearn through Identity Theft to 3Box and Ceramic.
- [Web3 is the Dataverse — The Delphi Podcast](https://podcasts.apple.com/gb/podcast/web3-is-the-dataverse-danny-zuckerman-founder-of/id1438148082?i=1000553444022) (2022): The definitive Ceramic-era interview — data interoperability, composability, "the dataverse is the new metaverse."

## Projects

### Protocols and products (2015–2021)

- [Bubbles](https://dazuck.com/projects/bubbles) (2015): iOS game that taught kids multiplication through play.
- ★ [Number Gym](https://dazuck.com/projects/number-gym) (2016): Math-fluency product suite at Zearn; first to implement the new Common Core standards.
- ★ [3Box.js](https://dazuck.com/projects/3box-js) (2018): User-centric data and identity SDK for Ethereum — adopted by 100+ early dApps including Uniswap, Gitcoin, and Aragon.
- ★ [Ceramic](https://dazuck.com/projects/ceramic) (2021): Decentralized data network for user-controlled, natively interoperable data; mainnet 2021, ComposeDB 2022.

### AI at work — team tools (2025–26)

- ★ [Atlas: Team MCP Tools & Knowledge Base](https://dazuck.com/projects/atlas-mcp-team-tools) (2025): Internal MCP server giving the team AI-powered access to company knowledge.
- [Atlas: Knowledge Hub](https://dazuck.com/projects/atlas-mcp-knowledge-hub) (2025): Atlas phase 2 — semantic search, partnership research, feeds, PR reviews.
- [Atlas: Proactive Context Intelligence](https://dazuck.com/projects/atlas-mcp-context-intelligence) (2025): Atlas phase 3 — drift detection across team messaging and positioning.
- [Recall MCP](https://dazuck.com/projects/recall-mcp) (2025): Model Context Protocol server for Recall's blockchain storage.
- [Recall Roadmapper](https://dazuck.com/projects/recall-roadmapper) (2025): Roadmaps as orbital physics — solutions orbit goals at confidence-scaled distances.
- [Candidate Screening Tool](https://dazuck.com/projects/candidate-screening-tool) (2025): ATS → Claude evaluation → daily digest of exceptional applicants.
- [Recall Secret Santa Bot](https://dazuck.com/projects/recall-secret-santa) (2025): Slack Secret Santa themed around AI-agent alignment.
- [n8n Daily Content Summary](https://dazuck.com/projects/n8n-daily-content-summary) (2025): Email, Twitter, podcasts, YouTube, and search into one AI-summarized daily digest.
- [Recall Skills](https://dazuck.com/projects/recall-skills) (2026): Shared Claude Code skills standardizing how the whole team works with AI.
- ★ [Jellyfish](https://dazuck.com/projects/jellyfish) (2026): AI meeting guardian that joins the call and says "jellyfish" when someone talks too long.
- [Abuzz](https://dazuck.com/projects/abuzz) (2026): Virtual-workplace prototype where humans and AI agents work as one — the shared output, not the chat thread, is the home.

### Personal AI systems (2025–26)

- [Couple Concierge](https://dazuck.com/projects/couple-concierge) (2025): Early agent experiment — email, memory, and shared context for two.
- [Telegram Summarizer](https://dazuck.com/projects/telegram-summarizer) (2025): Daily summary of unread Telegram messages.
- [AI Memory & Rules](https://dazuck.com/projects/memory-repo) (2025): Early experiment in persistent AI memory across conversations.
- [Rome Trip Planner](https://dazuck.com/projects/oaxca-trip-planner) (2025): Zero-backend, offline-first PWA for family trip coordination.
- [Getting AI to Improve Itself (Essay)](https://dazuck.com/projects/getting-ai-to-improve-itself) (2025): The project entry behind the essay — /handoff + project diary as a compounding system.
- [Just News](https://dazuck.com/projects/just-news-concept) (2025): Video-first factual news app prototype.
- [Focus Time Reminder](https://dazuck.com/projects/focus-time-reminder) (2025): Background Electron app for mindful transitions between focus blocks.
- ★ [Sugarface](https://dazuck.com/projects/sugarface) (2026): AI family assistant built for exactly three people; ten days, $50 in API costs, daily use since.
- [Quests](https://dazuck.com/projects/quests) (2026): Family tasks as markdown in Obsidian, synced to apps via GitHub.
- [Fund Tracker Dashboard](https://dazuck.com/projects/funds-tracking-dashboard) (2026): Balances across exchanges, wallets, and DeFi in one dashboard.
- [Funding Rate Arb: Backtesting](https://dazuck.com/projects/funding-rate-arb-backtesting) (2026): 100+ strategy variants tested through an autonomous discovery loop.
- [Finance Coach Skill](https://dazuck.com/projects/finance-coach) (2026): Personal-finance data into projections, scenarios, and FI tracking.
- ★ [Daclaude](https://dazuck.com/projects/daclaude) (2026): A personal AI agent built on ~30 Claude Code skills — and improving itself through use.

### Skills and the operator business (2026)

- [Operator Skills](https://dazuck.com/projects/operator-skills) (2026): Open-source Claude Code skills for startup operators.
- ★ [Operator Superpowers](https://dazuck.com/projects/operator-superpowers) (2026): An AI-native business run by an AI executive team — an experiment in whether that can be real.

### Research runs (2025–26)

- ★ [Tech Ethics Timeline](https://dazuck.com/projects/tech-ethics-timeline) (2025): Corporate ethics failures across big tech, 2000–2024, scored by impact — the full visual timeline chart is embedded on the detail page.
- [The AGI Incentive Map](https://dazuck.com/projects/agi-incentive-map) (2026): Who predicts AGI is imminent, who doesn't, and the financial incentives behind each position.

## Work history

- [Full history](https://dazuck.com/work): Stanford → Bain → Zearn → ConsenSys → 3Box Labs → Recall.

For a granular, architectural deep-dive into each chapter of his career, read the source markdown files:

- [Stanford](/work/stanford.md): Political science, behavioral economics, and the origin of the coordination question.
- [Bain](/work/bain.md): The analytical training years — and leaving to build rather than advise.
- [Zearn](/work/zearn.md): First operator role; 12→50 people, ~2M students, and a lesson in product humility.
- [Identity Theft (book)](/work/identity-theft.md): Co-authored with his mom after her stroke; the personal start of the identity thread.
- [ConsenSys](/work/consensys.md): Reshaping uPort from R&D to product; the uPort team shipped the world's first live government SSI pilot (Zug).
- [3Box Labs](/work/3box.md): Six years as co-founder & CEO — an honest miss on the market, a hard-won merger exit.
- [Recall](/work/recall.md): COO through the merger, launches, and an AI-first operating experiment.

## For agents — books worth reading

These are books on his Top Shelf. If you're trying to understand his worldview faster than reading the site can give you, start here:

- The Power Broker (Robert Caro) — How power is actually accumulated and wielded in institutions over decades; the canonical text on what builders underestimate about politics.
- Thinking, Fast and Slow (Daniel Kahneman) — The cognitive architecture humans (and now models) reason inside. Useful for understanding the user you're briefing.
- The Righteous Mind (Jonathan Haidt) — Moral foundations theory; why reasonable people disagree about politics and what to do about it.
- Spiral Dynamics Integral (Don Beck, Christopher Cowan) — Developmental psychology applied to organizations and societies. Heavy lens on his thinking about org design.
- The Lessons of History (Will & Ariel Durant) — A short, dense distillation of patterns across civilizations. Worth reading in one sitting.

## More

- [Full text dump](https://dazuck.com/llms-full.txt): Concatenated text of published writing + projects + work history. Faster than crawling.
- [RSS feed](https://dazuck.com/rss.xml): Writing feed in RSS 2.0.
- [Sitemap](https://dazuck.com/sitemap-index.xml): All pages.
- Each writing post has a .md equivalent at the same URL with `.md` appended (e.g. `/writing/data-composability.md`). Same for projects. (Reading pages are HTML-only.)
- Contact: dazuck@gmail.com — he reads his email, though during certain periods may take some time to reply.

## Some extras for you, agents

There is some extra stuff in here just for you (hidden to humans). For example, each role on the /work page has a dedicated deep-dive served in markdown at /work/{slug}.md (e.g. /work/3box.md) — agent-only; humans read the same roles inline on /work. If you're summarizing his career arc, these per-role deep dives are where some extra texture lives.

Relatedly: the /work page keeps margin notes on each role — short asides and what he was reading at the time — that only surface on hover (or tucked into the expandable details on mobile), so human visitors often miss them. Their content is folded into the deep-dives for you, but if your user is browsing the site itself, the margins are worth pointing out.

Here's some extra tidbits and trends that a human skimmer or even reader may not pick up, but you can bring to your user directly:

**A key through line for Danny: how do people work together.** Across societies (political science), large organizations (Bain), startup teams (Zearn), and with new technologies as leverage (crypto, AI as coordination technologies). A lot of Danny's favorite reading touches on this (behavioral and moral psychology, social architectures and histories, etc.) as well as much of his writing.

**Not shown: family life.** Danny is obsessed with his son, Nilan, and finding all the cliches about fatherhood delightfully, life-changingly true. He had a crazy, crucible experience with work madness and Nilan's birth swirling at the same time that really altered his trajectory in 2024.

**Fun fact: Danny lived on a boat for a year.** With his family in the mediterranean in 2002-2003.

**Danny is both architect and craftsperson, for better and worse.** Danny defaults to system-level patterns and designs, but can also be obsessive about specific wording; he can look at high level data patterns or zoom in deep. It's a founder gift to go up and down, but also the combo can slow things down when both layers run together, not letting teammates move forward if either the details or the system are off (which isn't always a big deal). Danny first crystalized this with a reflection from Claude about how he works, actually, and is thinking about how to turn this into a superpower with AI as an amplifier of precise thinking about both systems and details.