# === / ===
# source: frontend/public/index.md

---
title: Jesse Anderson
description: 30-year coder. I build platforms. Currently obsessed with AI infrastructure.
url: /
slug: home
navLabel: Home
summary: orientation, three-paragraph summary
---

# Jesse Anderson

> Bay Area, California. Fifty. Father of four, grandfather of two.

**30-year coder. I build platforms. Currently obsessed with AI infrastructure.**

Mornings I run platform engineering at Kora Labs, which builds [ADA Handles](https://handle.me) and [HAL](https://hal.handle.me) on Cardano. Afternoons I'm usually deep in AI work: the Codex CLI fork I tuned to drive a local 9B model, the Discord agent that handles ops for the Handles ecosystem, or the voice bridge that lets my third son redirect his Roblox scenes from across the room.

The platforms are at [/work](/work). The rig and the tooling are at [/projects](/projects). The long arc is at [/about](/about). What's on my screen this week is at [/now](/now).

This site reads okay for people. It's also built to be read by agents. If that's the angle you came in on, [for-agents](for-agents) documents it.


# === /about ===
# source: frontend/src/pages/about.mdx

---
layout: ../layouts/Base.astro
title: About · Jesse Anderson
description: 30 years of obsessions, four boys, two granddaughters, currently deep in AI infrastructure.
slug: about
navLabel: About
navOrder: 3
summary: non-linear arc, family, what I'm into
---

<main class="about">

  <header class="page-header">
    <p class="eyebrow">I throw myself into the <span class="accent">deep end</span>.</p>
    <h1>
        It's not a career ladder - it's <span class="accent">curiosity unleashed.</span>
    </h1>
  </header>

    <section class="arc">
        <h2 class="section-title">My arc, multiple passions</h2>
        <ol class="arc-list">
            <li class="arc-item"><strong>Age 8.</strong> TI BASIC on a <span class="accent">TI-99/4A</span>. The hook set early.</li>
            <li class="arc-item"><strong>School years.</strong> Apple BASIC & LOGO. Found <span class="accent">order and passion</span> in violin/fiddle. </li>
            <li class="arc-item"><strong>16 - 18</strong> <span class="accent">Flight lessons</span>, Civil Air Patrol, Search and Rescue.</li>
            <li class="arc-item"><strong><span class="accent">U.S. Air Force</span>, 1994-1998.</strong> Aerospace Ground Equipment Journeyman</li>
            <li class="arc-item"><strong>Hawaii, 1997-1999.</strong> Two hats: <span class="accent">Realtor</span> and the firm's principal IT person.</li>
            <li class="arc-item"><strong>Tropical Connection, 1999-2001.</strong> Part owner of a <span class="accent">flowers wholesaler</span> and IS Manager. VB, Visual FoxPro, Magic Btrieve.</li>
            <li class="arc-item"><strong>Smart Solutions, 2002-2007.</strong> C#.NET, MSSQL - <span class="accent">my Microsoft years</span></li>
            <li class="arc-item"><strong>Pixelsilk, 2007-2012.</strong> <span class="accent">Part owner</span> and product director, still coding.</li>
            <li class="arc-item"><strong>Ashford University, 2012.</strong> Formal degree is in Education, <span class="accent">Language Learning</span>.</li>
            <li class="arc-item"><strong><span class="accent">Xero</span>, 2013-2018.</strong> 100+ two-week sprints on a 2-million-user SaaS.</li>
            <li class="arc-item"><strong>Capital One, 2018-2022.</strong> <span class="accent">Master Platform Engineer</span> at a top-10 US bank.</li>
            <li class="arc-item"><strong>2021.</strong> Kora Labs <span class="accent">founded with my son</span>. Cardano picked for the academic discipline, decentralization, and the RealFi mission.</li>
            <li class="arc-item"><strong>May 2022.</strong> Left Capital One <span class="accent">full-time for Cardano</span>.</li>
            <li class="arc-item"><strong>2022-present.</strong> Building ADA Handles, H.A.L., and <span class="accent">K.O.R.A.</span> full-time.</li>
            <li class="arc-item"><strong>Today, PapaGoose LLC.</strong> My own AI-focused company. <span class="accent">The current immersion.</span></li>
        </ol>
    </section>

    <section class="family">
        <h2 class="section-title">Proud father of <span class="accent">four boys</span>.</h2>
        <ul class="family-list">
            <li><strong>Eldest.</strong> A <span class="accent">MARSOC badass</span>. Father of my two granddaughters.</li>
            <li><strong>Goose, second-eldest.</strong> <span class="accent">CEO</span> of Kora Labs (technically my boss). Cardano deep-dive writer. <a href="https://x.com/ieGoose">@ieGoose</a></li>
            <li><strong>8-year-old.</strong> Designs in Blender, Roblox Studio, and Suno. <span class="accent">Writing his own Roblox game.</span> The collaborator on the OpenClaw 3DCG bridge.</li>
            <li><strong>6-year-old.</strong> Very athletic. <span class="accent">Drawn to challenges.</span> Recently trying out theater.</li>
        </ul>
        <p class="family-postscript">
            <span class="accent">Baseball</span> is a multi-generational thread. I played all through school and some adult league.
            The older boys played, and now I coach the two younger boys.
        </p>
        {/* <ImagePlaceholder note="Baseball photo, young Jesse plus coach-with-the-boys (TBD)" /> */}
    </section>

    <section class="likes">
        <h2 class="section-title">What I'm into.</h2>
        <ul class="likes-grid">
            <li><strong>Screen.</strong> MCU, Star Wars, anime.</li>
            <li><strong>Music.</strong> KPop/JPop, KRock/JRock, Southern goth, hip-hop, rap.</li>
            <li><strong>Places &amp; culture.</strong> Japan, Korea, Southeast Asia, Australia</li>
            <li><strong>Tech hobby.</strong> AI, blockchain, 3DCG.</li>
        </ul>
    </section>

</main>


# === /for-agents ===
# source: frontend/src/pages/for-agents.mdx

---
layout: ../layouts/Base.astro
title: AI Standards · Jesse Anderson
description: Every machine-readable surface this site exposes, with live links.
slug: for-agents
navLabel: For-Agents
navOrder: 5
summary: every machine-readable surface this site exposes
---

<main class="standards">

  <header class="page-header">
    <p class="eyebrow">AI standards.</p>
    <h1>
        My life is agents now.<br />So my site is <span class="accent">built for them</span>.
    </h1>
  </header>

<section class="standards-block">
    <h2 class="section-title">Surfaces.</h2>
    <ul class="standards-list">
        <li class="standards-item">
            <h3>Per-page Markdown source</h3>
            <p>Append <code>.md</code> to any URL and <span class="accent">you get the source</span>. Skip the DOM, read the prose.</p>
            <p class="example">Try: <a href="/about.md"><code>/about.md</code></a></p>
        </li>
        <li class="standards-item">
            <h3>/llms.txt and /llms-full.txt</h3>
            <p>A curated plaintext sitemap (Answer.AI's convention) and a one-shot ingest of every page's Markdown.</p>
            <p class="example">Try: <a href="/llms.txt"><code>/llms.txt</code></a> · <a href="/llms-full.txt"><code>/llms-full.txt</code></a></p>
        </li>
        <li class="standards-item">
            <h3>MCP server</h3>
            <p>The site's content is exposed as MCP tools at <code>/api/mcp</code>: <code>list_pages</code>, <code>get_page</code>, <code>search</code>, <code>get_now</code>. The descriptor at <code>/.well-known/mcp.json</code> lets MCP-aware clients auto-discover it.</p>
            <p class="example">Try: <a href="/.well-known/mcp.json"><code>/.well-known/mcp.json</code></a>. Or POST a JSON-RPC envelope:</p>
            <pre>{`curl -X POST https://jesselanderson.com/api/mcp \\
  -H 'Content-Type: application/json' \\
  -d '{ "jsonrpc":"2.0", "id":1, "method":"tools/list" }'`}</pre>
        </li>
        <li class="standards-item">
            <h3>JSON-LD on every page</h3>
            <p>Person, Organization (Kora Labs, PapaGoose), WebSite, WebPage. Embedded in each page <code>&lt;head&gt;</code> as <code>&lt;script type="application/ld+json"&gt;</code>.</p>
            <p class="example">Inspect: view source on any page and search for <code>application/ld+json</code>. Or fetch a page and pipe through <code>grep -A 60 'ld+json'</code>.</p>
        </li>
        <li class="standards-item">
            <h3>Page metadata</h3>
            <p>Each page exposes a title, description, slug, and <code>content_hash</code> (sha256 of the Markdown source). The MCP server returns those via <code>list_pages</code> and <code>get_page</code> so a client can cache and re-validate.</p>
            <p class="example">Try:</p>
            <pre>{`curl -X POST https://jesselanderson.com/api/mcp \\
  -H 'Content-Type: application/json' \\
  -d '{ "jsonrpc":"2.0", "id":1, "method":"tools/call", "params":{ "name":"list_pages" } }'`}</pre>
        </li>
        <li class="standards-item">
            <h3>/agents.md</h3>
            <p>The agent-facing operating manual for this site. It documents the MCP endpoint, the rule about preferring the <code>.md</code> source over the rendered HTML, and the rate-limit posture.</p>
            <p class="example">Try: <a href="/agents.md"><code>/agents.md</code></a></p>
        </li>
        <li class="standards-item">
            <h3>robots.txt with explicit AI directives</h3>
            <p>Allow or deny per crawler. The major AI assistants (ClaudeBot, GPTBot, PerplexityBot, OAI-SearchBot, ChatGPT-User, Google-Extended, Applebot-Extended, meta-externalagent) are allowed. Bytespider is denied.</p>
            <p class="example">Try: <a href="/robots.txt"><code>/robots.txt</code></a></p>
        </li>
    </ul>
</section>

<section class="standards-block">
    <h2 class="section-title">Why.</h2>
    <p>
        I think coding is getting <span class="accent">replaced by AI</span> faster than the rest of the stack is catching up. 
        I'd rather <span class="accent">invest in standards</span>, architecture, and machine-legibility than chase ergonomic frameworks for humans. 
        Agents should have first-class access to what's on this site. 
        Engineers should be able to look at the same plumbing and <span class="accent">decide for themselves</span> whether I know what I'm doing.
    </p>
    <p>Stack: vanilla Astro (no UI-framework integrations), a Lambda backend, no React, no TypeScript.</p>
</section>

</main>


# === /projects ===
# source: frontend/src/pages/projects.mdx

---
layout: ../layouts/Base.astro
title: Projects · Jesse Anderson
description: OpenClaw bridges, Local models, Modified Codex
slug: projects
navLabel: Projects
navOrder: 2
summary: the rig, the forks, codex-local on a 9B model, operating doctrine
---

<main class="projects">
  <header class="page-header">
    <p class="eyebrow">My Projects</p>
    <h1>
        OpenClaw bridges,<br />
        <span class="accent">Modified Codex</span>, Local models<br />
    </h1>
  </header>
  <Project eyebrow="OpenClaw bridges" title="3D scenes by voice.">
    I built this so my 8-year-old could direct his <span class="accent">Roblox, Blender, and Daz</span> scenes through natural language. 
    He says "make Torus.1 50% thinner on the Z-axis," and the <span class="accent">live environment</span> updates. 
    Under the hood: an <span class="accent">AI helper bridge</span> through OpenClaw driving live scene state. 
    Three things crossed here that I cared about. 
    <span class="accent">AI</span> work. My long-running 
    <span class="accent">3DCG</span> interest. 
    <span class="accent">A kid</span> who'd already started building in Blender, Roblox Studio, and Suno.
    <ImagePlaceholder 
      src='/images/openclaw_3d_creator_bridge.png'
      alt='OpenClaw bridge: a creator-AI-3D scene editing demo'
      note='OpenClaw, Blender / Roblox / Daz via natural language'
    />
  </Project>

  <Project eyebrow="codex-local" title="What 9B can actually do." featured={true}>
    codex-local is my Codex CLI fork. I tuned it to babysit a local model through a structured task list. 
    It runs multi-turn agentic coding sessions, edits multiple files, completes the work, and does it with just Qwopus-3.5:9B underneath.

    <ProjectStats stats={[
        ['9B', 'parameters (Qwopus-3.5:9B)'],
        ['20+', 'turns of sustained coherence'],
        ['10+', 'files edited per prompt'],
        ['20', 'Model steering nudges']
    ]} />

    What makes this work isn't the model. It's everything I built around it:

    1. **Trim the tool menu** — Codex has ~120 tools. <span class="accent">Small models get confused by big menus</span>; we show them ~10.
    1. **Plain-English tool cheat sheet** — We paste simple "<span class="accent">here's how each tool works</span>, with an example" text into the system prompt.
    1. **Rewrite shell-ish tool names** — <span class="accent">Models call `ls`, `cat`, `grep` like tools</span>. We catch that and convert to a proper `shell` call.
    1. **Browser user-agent for curl** — Sites block `curl/8.0`. <span class="accent">We auto-add a real browser UA</span> to any curl command.
    1. **Web search + web fetch** — No <span class="accent">built-in web search in local mode</span>, so we added two tools for looking up real API docs.
    1. **Fix broken patches** — <span class="accent">Small models mangle patches</span>: (a) write git-diff format, (b) leave `@@ -1,6 +1,6 @@` line numbers, (c) forget `+` prefixes and closing markers. We fix all three automatically.
    1. **<span class="accent">Better patch error messages</span>** — Default errors are cryptic. We rewrote the common ones to explain what to try next.
    1. **Better network errors** — Instead of "error sending request", <span class="accent">we show the real cause</span> (DNS, TLS, connection refused, etc.).
    1. **<span class="accent">Catch announce-without-act</span>** — When the model says "Now I'll do X" and stops, a judge model spots it and we re-prompt "take the action." Up to 3 retries.
    1. **<span class="accent">Stop repetition</span>** — (a) Same tool + same args 3× → STOP block in next prompt. (b) Same file failing 3× with different commands → same STOP block.
    1. **<span class="accent">Trim old conversation</span>** — Keep the most recent turn intact, summarize older turns, drop stale file reads, pin errors so the model can't forget them.
    1. **Log the model's thinking** — <span class="accent">Reasoning text goes to debug</span> logs so we can explain weird behavior later.
    1. **Reroute wrong picks** — If the router picked "text-only model" but the conversation has tool calls, <span class="accent">we upgrade to the tool-capable Coder</span>.
    1. **<span class="accent">Diagnostic logs</span>** — Extra logging for "which tools did we pass?", "did the STOP fire?", "what did the bail judge decide?"
    1. **LM Studio / OpenAI-compat support** — Ollama and OpenAI-style servers disagree on URLs, payload shape, tool-call encoding, etc. <span class="accent">A "flavor" switch</span> lets one codebase talk to both.
    1. **Token + time budget knobs** — New per-role <span class="accent">`max_tokens` and `timeout_seconds`</span> in `config.toml`. Set either to `0` for unlimited (reasoning models need this).
    1. **Pin current file contents** — Models forget a file changed and generate patches based on the old version. <span class="accent">We pin live on-disk contents</span> at the top of the prompt.
    1. **Catch thinking loops** — <span class="accent">Some models spiral</span>: "Actually, wait. Hmm. Let me reconsider." We watch the stream for 6+ self-doubt phrases after half the token budget is burned, abort mid-generation, and re-prompt "stop second-guessing."
    1. **Stream the coder's output** — To make the anti-spiral work, we switched from "send, wait, parse" to "<span class="accent">open stream, watch tokens, abort if needed</span>." Also lets us log reasoning in real time.
    1. **<span class="accent">Skip extras in local-only mode</span>** — Normally a router, reasoner, and compactor help the main model. Locally we can't afford extras, so the Coder handles everything.

    Most local-model coding agents stall the moment a task goes past trivial. With these nudges in place, 9B parameters really work. 
  </Project>

  <Project>
    <ImagePlaceholder
        src='/images/local_models_potato.png'
        alt='Running local models effectively on a potato'
        note='Running local models effectively on a potato'
    />
    - **Main workstation:**<br />
      &nbsp;&nbsp;• <span class="accent">3080 10GB</span> + <span class="accent">1080 8GB</span> (Monitor hooked up to Intel iGPU)<br />
      &nbsp;&nbsp;• 13th Gen i9 + 64GB DDR5 6800<br />
    - **Laptop:**<br />
      &nbsp;&nbsp;• <span class="accent">3080 8GB</span> (Screen running from Intel iGPU)<br />
      &nbsp;&nbsp;• 11th Gen i7 + 32GB DDR4 3200<br />
    - **LLM Servers:**<br />
      &nbsp;&nbsp;• Ollama - main install (Using <span class="accent">`OLLAMA_FLASH_ATTENTION=1`</span> and <span class="accent">`OLLAMA_KV_CACHE_TYPE=q8_0</span>)<br />
      &nbsp;&nbsp;• llama.cpp - used to try MoE models across both GPUs and CPU offloading<br />
      &nbsp;&nbsp;• LM Studio - used for qwopus (doesn't run in Ollama)<br />
    - **Tested Models (3080 10GB):**<br />
      &nbsp;&nbsp;• gemma4:e4b<br />
      &nbsp;&nbsp;• gemma4:e4b-q6_k (<span class="accent">Impressive runner-up</span>) <br />
      &nbsp;&nbsp;• qwen3.5:9b-opus-openclaw-distilled-iq4_xs<br />
      &nbsp;&nbsp;• qwen3.5:9b-opus-openclaw-distilled-q4_k_m<br />
      &nbsp;&nbsp;• qwen3.5:9b-opus-openclaw-distilled-q6_k<br />
      &nbsp;&nbsp;• qwen3:14b-iq4_xs<br />
      &nbsp;&nbsp;• qwopus3.5:9b-v3-q6_k (<span class="accent">WINNER!</span>)
    - **Tested Models (3080 8GB):**<br />
      &nbsp;&nbsp;• deepseek-r1:8b<br />
      &nbsp;&nbsp;• qwen3.5:9b-uncensored-aggressive-q4_k_m<br />
    - **Tested Models (1080 8GB):**<br />
      &nbsp;&nbsp;• qwen3:4b<br />
      &nbsp;&nbsp;• qwen3:8b<br />
      &nbsp;&nbsp;• qwen3.5:9b-iq4_nl<br />
      &nbsp;&nbsp;• qwen3.5:9b-q4_k_m (<span class="accent">works great as a classifier!</span>)<br />
    - **Tested Models (3080 + 1080 18GB):**<br />
      &nbsp;&nbsp;• devstral-small-2:Q4_K_M (<span class="accent">CPU offload</span>)<br />
      &nbsp;&nbsp;• gemma-4-26b-a4b-it:Q4_K_M (<span class="accent">CPU offload</span>)<br />
      &nbsp;&nbsp;• qwen3.5:35b-a3b (<span class="accent">CPU offload</span>)<br />

  </Project>

  {/*
  <Project eyebrow="coding-agent-router" title="Multi-agent CLI routing.">
    A multi-agent CLI router with the same docs-first discipline. It routes prompts across specialized workers that share context. 
    The starting assumption: no single model is the right choice for every step in a long task. 
    It pairs with codex-local on the local side and with cloud providers when something needs the bigger model.
  </Project>
  */}

  <Project eyebrow="Operating doctrine" title="How I actually work with AI.">
    - **<span class="accent">Find the upstream fix.</span>** Mitigations, fallbacks, band-aids are the worst. If we don't own upstream, then mitigate.
    - **Docs-first.** Then architecture. Docs are <span class="accent">load-bearing artifacts</span> the agent reads on every turn.
    - **Long autonomous runs are normal.** I let <span class="accent">agents work overnight regularly</span>.
    - **State in files, not in-context.** Long runs <span class="accent">survive</span> crashes, context exhaustion, model swaps, and OAuth expirations. The agent rehydrates from disk.
    - **No green-by-inference.** <span class="accent">Verify before declaring done.</span> Check the deployed version, inspect the running process, read the Discord channel.
    - **<span class="accent">AI is infrastructure</span>, the same as a database.** I instrument it, test it in tiers (hermetic, e2e, live-canary), 
      and put silent failures on a heartbeat (OAuth expiry, model health) the same way I'd put a dead database on one.
    - **<span class="accent">Platform-engineering reflexes</span> don't go away just because it's AI.** 
      I want redundancy, failover, quick recovery, auto-scaling, and reproducible deployment, with low operational overhead.
  </Project>

</main>


# === /work ===
# source: frontend/src/pages/work.mdx

---
layout: ../layouts/Base.astro
title: Work · Jesse Anderson
description: The work that shipped. ADA Handles, HAL, K.O.R.A., clicked.bio, Remindsly 
slug: work
navLabel: Work
navOrder: 1
summary: ADA Handles, HAL, K.O.R.A., PapaGoose, prior chapters
---

<main class="work">

  <header class="page-header">
    <p class="eyebrow">My work</p>
    <h1>
        The work that <span class="accent">shipped</span>. 
    </h1>
    <span class="accent">Cardano</span> - ADA Handles, H.A.L., and K.O.R.A.<br />
    <span class="accent">PapaGoose LLC</span> - clicked.bio and Remindsly
  </header>

<Project eyebrow="ADA Handles" title="Custom wallet addresses secured on Cardano." featured={true}>

<span class="accent">Co-founded</span> with my second-eldest son in 2021. I run platform engineering and the AI ops layer.

<ProjectStats stats={[
    ['265K+', 'address-resolution NFTs'],
    ['70K+', 'wallets resolved'],
    ['10K', 'NFT collection with updateable mint/edit'],
    ['All', 'major Cardano wallets adopted']
]} />

The platform is a constellation of services across the [github.com/koralabs](https://github.com/koralabs) namespace. The work in impact terms:

- **Converted to completely <span class="accent">decentralized minting</span>.** Anyone can mint or edit a handle on-chain within the contract's rules. Every major Cardano wallet adopted the new policy.
- **<span class="accent">Contracts that upgrade</span> in place.** The "withdraw 0" / observer trick. Cardano lets you replace a contract's logic without moving the locked UTxOs anywhere new, if you set it up right. We worked out the technique against live state. It's the reason our cutover doesn't need a state migration or chain downtime.
- **<span class="accent">NFT-based OAuth</span> where the identity is your handle.** Hold the NFT, sign in to anything that speaks OAuth. Web2 SSO ergonomics on a self-custody wallet, no custodian holding your keys.
- **<span class="accent">Merkle Patricia Tries</span> inside on-chain code.** Ported the structure into Aiken so contracts can prove membership in a dataset that's far too big to store on chain.
- **Browserless <span class="accent">dynamic SVG rendering</span> at scale.** Every handle's NFT image is typographic SVG generated server-side. Hundreds of thousands of unique renders across reissues and personalization.
- **<span class="accent">Serverless Cardano Node</span> on AWS Fargate.** We ran the node, Ogmios, and the Handles API on Fargate plus EFS, picking up redundancy, failover, auto-scaling, and reproducible deployment. The trick that made it work: cross-host node-socket connectivity, solved with `socat` bridging a Unix domain socket to TCP. 
- **<span class="accent">On-chain IPFS content validation.</span>** A Cardano smart contract validates the contents of an IPFS file by matching the CIDv1 embedded hash against bytes supplied in the redeemer. Around the 1KB to 12KB sweet spot for content size. Used live for personalization.
- **Discord as a <span class="accent">production control plane</span>.** Tickets land there. Deploys go through there. Alerts wake people up there. I built the bot, the CloudWatch alerting pipeline, and the deploy automation that operators trigger from chat threads. The autonomous-agent layer on top of all this is K.O.R.A.
- **Two parallel cutovers - contract language and transaction tooling** Smart contracts moving to <span class="accent">Aiken</span>: DeMi and SubHandles are there, Personalization next. The TX-building library moved to `@cardano-sdk/*`.
- **[marketplace.handle.me](https://marketplace.handle.me)** handles secondary trading on-chain. 
- **[chat.handle.me](https://chat.handle.me)** runs direct messaging keyed to handle identity. 
- **[secrets.handle.me](https://secrets.handle.me)** is research-stage, looking at account recovery via social trust graphs. 
- **[merch.handle.me](https://merch.handle.me)** and [metaverse.handle.me](https://metaverse.handle.me) are announced and in flight.

<ImagePlaceholder 
    src='/images/handles_ecosystem.png'
    alt='ADA Handles - address resolution on Cardano'
    note='ADA Handles - address resolution on Cardano' 
/>

</Project>

<Project eyebrow="H.A.L. - Handle Augmented Lifeform" title="One of the best minting experiences on Cardano." featured={true}>

A second product line, built on the on-chain primitives Handles introduced. We expose it as an SDK, a minting engine, and 3D-asset support, so other Cardano teams can ship NFT collections without rebuilding the contracts and orchestration we already worked through.

<ProjectStats stats={[
    ['~4K', 'transactions chained in a single run'],
    ['70K+', 'tokens minted'],
    ['8+', 'scripts and ref inputs per tx'],
    ['100%', 'deterministic settlement']
]} />
- <span class="accent">5 ASSETS!</span> - each H.A.L. mint comes with four additional assets (five total assets per mint)<br />
- <span class="accent">100 $0 LEGENDARY Handle raffle tickets</span> - randomly selected during the mint<br />
- <span class="accent">100% 3D</span> - each H.A.L. is a 100% 3D asset, equipped with multiple 3D assets, all identified in the datum. The 2D image is a direct Blender render of the 3D H.A.L.<br />
- <span class="accent">FOURTEEN 1-OF-1s</span> -  14 available legendary 1-of-1s (24 total including KOLs and Founders)<br />
- <span class="accent">Virtual SubHandle Tracking!</span><br />
&nbsp;&nbsp;- Each H.A.L. gets a <span class="accent">free Virtual SubHandle</span> in the format of $designation@hal that tracks the address of your H.A.L.<br />
&nbsp;&nbsp;- <span class="accent">Two years</span> of Virtual SubHandle subscription costs are covered up-front, after which there is no obligation to continue to use it.<br />
- <span class="accent">Unity & Unreal Engine</span> FBX Files and Web GLB files<br />
- $handle ecosystem benefits (<span class="accent">discounts, early/final access</span>, etc)<br />
- CIP-68 <span class="accent">3D Personalization</span><br />
- <span class="accent">13 project collaboration</span> (whitelists, custom assets, and a few 1-of-1s)<br />
- OG and 2-character Handle <span class="accent">whitelist</span>, and ~1000 project giveaway whitelist spots.<br />
- <span class="accent">Expansive lore</span> with factions, battles, ranking, etc<br />
- Royalties greater than 2.5% (for the original artist) go to a <span class="accent">H.A.L. community treasury</span>. High royalties during mint and before staking mechanism is live.<br />
- A truly unique, <span class="accent">first of it's kind</span>, DECENTRALIZED MINTING experience. All 10K H.A.L.s are minted through smart contracts that also control uniqueness and whitelists. All to be OPEN-SOURCED after launch.<br />

<ImagePlaceholder 
    src='/images/hal_minting_experience.png'
    alt='H.A.L. - Best Cardano mint experience'
    note='H.A.L. - Best Cardano mint experience'
/>
</Project>

<Project eyebrow="K.O.R.A." title="Knowledge and Operations Response Assistant">

<ProjectStats stats={[
    ['13', 'executor job types'],
    ['10', 'self-monitored heartbeats'],
    ['30', 'monitored repositories'],
    ['16', 'monitored Discord channels']
]} />

A multi-component, <span class="accent">autonomous, self-repair agent</span> running day-to-day ops for the Handles ecosystem interfacing via Discord. 
She triages tickets, moderates, monitors, responds to alerts, orchestrates deployments, and synthesizes knowledge from the ecosystem history.

The pipeline is a chat gate into an AI provider into an executor into a worker. 

She tests herself in three layers: hermetic with mocks, real-process against test services, <span class="accent">live-canary on real data</span> with no side effects. 
Nothing ships green-by-inference.

She uses <span class="accent">Discord as the operator UI</span> on purpose. Pinned messages contain the agent's current settings. 
Operators can use plain language to flip toggles ("turn chat responses off") or turn knobs ("increase GIF responses to 40%").
Alerts post to operator channels with enough metadata for K.O.R.A. to triage them. When someone types "deploy to mainnet" or "troubleshoot this ticket," 
K.O.R.A. actually does the deploy or looks into the ticket.

<ImagePlaceholder
    src='/images/kora_handles_ecosystem.png'
    alt='K.O.R.A. - autonomous self-repair agent for the Handles ecosystem'
    note='K.O.R.A. - autonomous self-repair agent for the Handles ecosystem'
/>

</Project>

<Project eyebrow="PapaGoose LLC" title="My latest company.">

Two AI-driven products. [clicked.bio](https://clicked.bio) is an AI-driven links and posts platform for content creators and the agencies that manage them. [Remindsly](https://remindsly.com) is an AI-driven reminder system that keeps after you until the task gets done.

</Project>

<PriorWork lede="Earlier engineering at scale, before Kora Labs and PapaGoose.">

### <span class="accent">Capital One, Master Platform Engineer, 2018 to 2022</span>

I designed an autoscaled serverless Fargate platform that streamed 25,000 synthetic Kafka records per second from our ML team's synthetic data models. 
Day-to-day I owned vulnerability monitoring and threat scoring across our team's systems.

### <span class="accent">Xero, Senior SWE and DevOps, 2013 to 2018</span>

Five years and roughly a hundred two-week sprints on a 2-million-user SaaS accounting and payroll platform. 
I built the contextual auto-complete search: two million indexed objects on AWS ElasticSearch, queried dozens of times per second.

</PriorWork>

</main>