AC Andrew Ciervo
12:00PMSouthwest Florida
Portfolio · Punta Gorda, Florida

Andrew
Ciervo

Electronics tech turned builder. I make the systems I actually want to live with — a house that tells me who just pulled into the driveway, a car app that clears fog off the map as I drive, and a photo pipeline my coworkers use every other Sunday. This page shows what they look like, how they work, and what broke along the way.

17things built and shipped
85,000+lines of code across them
15years fixing electronics & helping people
The screen in our living room waking up: the clock fades, the camera swoops down onto a 3D model of our real house and lot, then the cards slide in one at a time. It runs on a Raspberry Pi I already owned, the same one doing the camera AI. How it works ↓

Everything, at a glance

Click any card to jump to its write-up. The colored tag says where it lives: public (open it right now), private (runs for me or a few people), or playable (try it on this page).

Private · daily use
Cape HouseCamera AI, voice announcements and a 3D home screen, all on one Raspberry Pi.
Python · three.js2026
Public beta
UnchartedA Tesla-browser app: every road you drive clears the fog of war.
WebGL · MapLibre2026
Public · in use
Pulte Pic SystemReads a marked-up PDF map, plans the route, shoots and delivers construction photos.
OpenCV · Apps Script2026
Public
MONSOON S13A wiper/washer controller board for a 1993 240SX, with a browser test bench.
KiCad · ATtiny16162026
Private · retired
Blockchain SurferDeFi scanner with a rules engine and an atomic flash-loan contract on Base.
FastAPI · Solidity2026
Private · in development
Property OSCRM, construction tracking and deals for a real-estate office, multi-tenant.
Next.js · FastAPI · Supabase2026
Private
TeslaDashLive car data through Tesla's Fleet API, with gauges and a rank system.
Node · WebSocket2026
Playable
Dash LED buildESP32 light strip for a Model 3 dash. No wires cut. Supercaps for the goodbye.
ESP32 · WLED usermod2026
RISK 6% → 2% · 1 PER SYMBOLPrivate
AI MT5 botForex bot I rebuilt after reading my own trade log for the real cause of losses.
Python · MetaTrader 52026
Playable
House BuilderA browser game: build a homestead on 1–20 acres, with a drone that delivers what you order.
three.js2026
45 FT · GABLE 8/12 · UE 5.8Private
Barndominium walkthroughOur future house, modeled in Fusion 360 and walked through in Unreal with a PS5 controller.
Fusion 360 · Unreal 52026
Public · client
New England Pressure CleaningA real business's website in Naples, FL. Fast, static, no dependencies.
HTML · CSS2026
Public · concept
Finance firm redesignAn unsolicited redesign concept, built as a pitch.
HTML · CSS2026
Public
Our Big MoveThe family homestead plan, with a floor plan drawn true to a hand sketch.
HTML · CSS2026
Private
FL Land HunterOne-file offline tracker for 5–10 acre parcels: flood zone X, paved road, under $100k.
HTML2026
Spin it
Charging dockA 3D-printed desk dock written as a parametric Fusion 360 script.
Fusion 360 API · Python2026
00 00 00 00 6D 64 61 74
size = 0  → never finalized
patch → moov_start − 32
HEVC → H.264  ✓ plays
One-off
Dashcam recoveryGot crash footage off a corrupted card for a neighbor's insurance claim.
MP4 forensics · ffmpeg2026
01 Cape House · Home AI · Raspberry Pi · Private, used every day

A house that knows who just pulled in

Jenna and I are moving into a new house in Cape Coral. The ISP's router does the bare minimum and every smart camera wants a subscription. So I built our own system on the Raspberry Pi already sitting on my desk — budget: $0.

It watches the cameras, works out what's actually happening ("Jenna's Model 3 just pulled into the garage", "FedEx delivery at the front door"), says it out loud in a natural voice, and shows it all on a 3D model of the house on our TV. It's also the start of the firewall and VPN for the whole house.

PrivatePythonDocker ComposeFrigateONNX RuntimeCLIP · YOLOX · YuNet · PaddleOCRthree.jsMQTTnginxTailscaleHome Assistant
Hub home screen: a 3D model of the house with recent activity, garage status, lights, thermostat and camera tiles Security screen with live camera and activity list Scene viewer showing a red Tesla outlined with confidence bars AI training tab with a grid of detections Network screen with protection layers and gauges Settings screen with voice announcement options Clock screen

Spoken announcements. A chime, then the voice; each word lights up as it's said, and the 3D view turns toward the spot. Urgent ones are casual on purpose: “Yo, someone's walking into the garage.”
The wake-up flight, staged one step at a time — flight, then frame, then cards, then camera cones — because the Pi's GPU stutters when everything moves at once. Shadows are baked, never drawn live.

What the cameras saw

Real scenes from the test camera, as the Hub shows them. The AI draws its own outline on the sharp screenshot, with how sure it is. (A delivery driver's face is blurred here; visitors' names are swapped for generic words.)

White Tesla pulling out of the garage, outlined at 99%
“Jenna's Model 3 Long Range pulled out of the garage and left.” Known cars need color + verified photos + a body-shape check.
Delivery driver walking up the driveway
“FedEx delivery in front of the house.” 40 sharp screenshots; uniform + walk to the door + truck at the curb.
Person rolling a bin at dawn
“Jenna took the trash out.” Someone who starts at the garage is from the house — never a delivery, visitor or alert.

From tire to voice, in about four seconds

Press play to walk through what happens when a car pulls in. Each stage is a real piece of the system, with the timings I measured on the Pi.

0.00s

Typical timings, stage by stage. Night, glare and palm fronds in the way make it slower; when the AI isn't sure yet, the screen says “Checking what drove by…” and stays quiet instead of guessing out loud.

What's inside the Pi

Everything runs in Docker on a 4 GB Raspberry Pi Compute Module 5. Dashed red boxes are services that have no login, so they never get a port on the network. The moving lines are data.

RASPBERRY PI CM5 · 4 GB · DOCKER COMPOSE Cameras2560×1920 main · 640×480 sub Reolink NVRRTSP · view-only account go2rtc restream:1984 never published Frigate 0.18detect 4 fps · zones · faces:5000 never published Mosquitto (MQTT)localhost + Docker net only hub-apiPython · ~6k lines · SQLite (WAL) live.py ······ MQTT feed, paths, zones grabber ····· full-res keyframes → RAM detector ···· YOLOX-nano / YOLOX-s describer ··· CLIP + nearest examples faces ······· YuNet + video refiner lpr / looker PaddleOCR plates + logos memory ······ vehicle fingerprints scenes ······ groups + one sentence announce ···· what to say, and when cameras ····· AI on/off over MQTT Only reachable through nginx /api/.Every POST needs the Hub PIN. Kokoro voicenatural, pre-recorded Piper voicequick fallback nginx :80allowlist: localhost,house subnet, Tailscaleread-only proxy paths Home Assistantlights · thermostat · plugs The HubTV kiosk · PC · tablet Tailscalephones, away from home no ports forwardedon the router

Locks on the doors

The part most people never ask about. A camera system is a set of eyes on your family — so the defaults are “closed” and every write needs a reason.

Nothing without a login touches the network

Frigate's API (:5000) and the go2rtc restream (:1984) have no authentication, so they're never published. nginx proxies only the specific read-only paths the Hub needs: snapshots, events, stats and the live socket.

An allowlist at the front door

nginx answers only localhost and Docker, the house subnet (set per install — nothing depends on today's Wi-Fi or IPs) and Tailscale's 100.64.0.0/10. Everyone else gets nothing.

A PIN on every change

Reads are open inside the house; writes are not. Every POST — labeling a face, turning a camera's AI off, teaching it a car — needs an X-Hub-PIN header checked on the Pi. The PIN lives in a 600-permission secrets file, never in the page.

Remote access without opening the router

Phones reach home over Tailscale (WireGuard). Zero port forwards. The Pi also advertises an exit node, so a laptop on hotel Wi-Fi can send everything home first.

Least privilege for the cameras

The Pi logs in to the recorder with a view-only guest account. It can watch; it can't change a setting or delete a recording. RTMP stays off.

Boring server hygiene

SSH by key only (passwords off), rpcbind masked, unattended security upgrades, and every container capped so the AI can't starve the TV or the voice.

Changes that can't lock me out

Any firewall or network change ships with a timer that rolls it back unless I confirm SSH still works. Every system file is copied to file.bak-YYYYMMDD before it's touched.

Built here, installed there

It was built on a test network and moves to the new house later, so nothing is hard-coded to a Wi-Fi name, subnet or IP. The camera zones get re-measured on site with a script.

Teaching it, one Saturday at a time

The describer started as a generic image model guessing from a few words. Each round I hand-checked every detection from the camera, and it learned from my labels. The test is leave-one-out over everything labeled so far, so later rounds are harder exams.

All answersOnly when it says it's ≥ 80% sure
Show as a table

When it says it's sure, it's right 95% of the time.

That's the number that matters for a house that talks: 262 of 276 confident answers were right in the last round. When it isn't sure, it waits for another look instead of announcing a guess. Overall it went from 50% to 83% on a test eight times bigger.

  • ~3,000 examples: my own labels (weighted ×3–4) plus Open Images crops for rare vehicles.
  • Pooled looks: each detection combines every sharp screenshot it's in, counted as at most 3 independent looks.
  • Tuned with a grid search over k, blend weight and class balance, using the same leave-one-out test.

Smooth on a Raspberry Pi

The Pi's GPU is the bottleneck, and I watch the TV closely. Every animation change gets measured on the real TV in frames per second before I call it done.

58fps idle, 3D home
55fps during the arrival flight
44fps dragging the house around
11fps when the TV woke up at 4K (fixed)
Baked, not live

Live shadows cost 12–15 fps while turning. Now the shadow map is drawn only when the sun moves, into a 1024² top-down mask the ground reads. The sky is one still picture, redrawn every 5 minutes.

Quiet while it moves

Background AI steals memory bandwidth from the GPU. Before any big animation the Hub calls /api/quiet, which freezes the voice recorder (SIGSTOP) and pauses the describer between frames.

Lighter in the background

Idle 3D redraws at 15 fps, 24 with labels up — the kiosk's idle CPU went from ~59% to ~37%. Lists update in place by key, so nothing re-animates or flickers on refresh.

What broke, and what I changed

The honest part. Every one of these happened on the live system, and each one turned into a rule.

  1. Unplugging one camera froze all of them.

    The recorder's RTSP server hung for every channel while go2rtc kept retrying the dead one. Now a removed camera's streams are removed too.

  2. The “sharp” screenshots showed the street before the car got there.

    I'd assumed 0.3 s of stream delay. Matching AI outlines to the live paths showed 3.6 s — two of those seconds were ffmpeg holding each keyframe until the next. Low-delay flags fixed it; the real figure is 1.6 s and every new camera gets measured.

  3. A 1 GB vision model killed the TV.

    Testing Florence-2 pushed the Pi out of memory and the kernel shut down the TV's browser mid-use. Rule now: small models only — the text finder I ended up with is 4.7 MB.

  4. The AI learned the parking spot instead of the car.

    Color examples taught it that anything in my spot is black. Our own cars now train a separate color model, and when the pixels and the describer agree, they win (my car went from 14/73 to 73/73 “red”).

  5. “Andrew's car left” at 5 and 7 PM. It hadn't.

    Frigate's box jumped from a car driving by onto my parked one. Jumping tracks are detected now, and a car found again where another one was parked is the same car.

  6. The TV woke up at 4K.

    Turning the TV off and on doesn't unplug it, so the display tool never re-applied 1080p: everything half size, 11 fps. A small guard checks every 3 seconds and puts it back.

  7. An upgrade erased history it couldn't redo.

    A model version bump cleared descriptions for 129 detections whose screenshots had already been trimmed. Restored from a database copy; now only detections that can be re-described are ever cleared.

02 Uncharted · Web app for the Tesla browser · Public beta

Every road you drive clears the fog

The map starts dark. Every road you drive reveals it for good, so your commute slowly becomes a map you uncovered. On the way it tells you about the places you pass, scores how smooth your driving is with a virtual glass of water, and flashes red-blue when Florida 511 reports police activity nearby.

Hard rule: it runs in a car's browser with no build step, no accounts, no API keys and no paid services.

A simulated drive through Boston, recorded from the live site. The places, photos and stories come from Wikipedia as you go.

Try the fog

A small version of the idea, running right here. Move your mouse or finger to drive — or leave it alone and it drives itself. It keeps track of the ground you've uncovered, like the real app's grid.

0.00 km²explored
0%of this town
move to drive · it drives itself when you stop
The real one does this on the GPU for 40,000 cells at once.
Car screen with energy and Superchargers
Car screen. A live Model 3 with g-forces, an energy model for the trim it detects from the browser, and the nearest Superchargers.
Star map
Sky tab. 5,044 real stars oriented to where the car points, the planets, the Moon's real phase. Unlocks for passengers or when parked.
Collection of places
Collection. Places you've passed, with rarities, quests and XP. It locks while moving: eyes on the road.
The fog: 800 ms → under 1 ms

The first version projected each explored cell every frame — 100–800 ms per frame at 40,000 cells. Now it's a WebGL custom layer: one 1024² texture aligned to the map, re-uploaded only when new ground is revealed (profiled at 0.45 ms). Pan, zoom and tilt cost nothing.

A voice that works in a car

Linux Chromium in the car often has no speech voices at all. So it ships a neural voice (Piper) that runs in a background worker, entirely in the browser — a one-time 63 MB download, then it reads stories offline.

Free data only

Map tiles from CARTO/OpenStreetMap, places from Wikipedia, weather from Open-Meteo, cameras from Overpass, incidents from FDOT's FL511 feed, alerts from the National Weather Service. No keys to leak, nothing to pay. 80 browser tests run against the live site.

03 Pulte Pic System · Field app + agents' site · Public, used every other Sunday

It reads the map, plans the drive, and delivers the photos

Every other Sunday I photograph new-construction lots in a community in Estero so three sales agents can send each buyer a progress update on their home. The agents mark the lots on a PDF map with colored stars. It used to be a morning of squinting at the map, driving in circles and sorting photos by hand.

Now my phone reads their stars straight out of the PDF, plans a route that goes around the lakes, walks me lot to lot, and the photos land in each agent's page in full resolution before I'm back in the truck. Running cost: $0.

Pulte Pics phone app: new photo run, upload the starred map
datbluepony.github.io/pulte-pics/about.html
Pulte Pics about page
The showcase page I wrote for the agents. Real numbers: 2,400 lots on the master map, ~1 minute to read every star, photos live while I walk to the next house, 13 days to download.

Plan a run

A live, simplified version of the route planner: stars land on lots, then it visits them over the streets only — never across the lake. It's a real breadth-first search on a grid, same idea as the app's rasterized street map.

0lots starred
0 mroute over streets
Blue, pink and yellow stars = three agents.
01 · Read the stars

pdf.js pulls each star's position and color (the color is the agent) from the PDF's vector drawing — not from pixels — plus the legend names.

02 · Find the lot

OpenCV template matching finds where each map screenshot sits on the master community map; each star gets the nearest lot label, checked against a lot index.

03 · Drive & shoot

Compass, distance and mini-map to the next lot. Its own camera: lens choice, Night mode for interiors, rotation baked into the file so photos are always upright.

04 · Deliver

Google Apps Script in my own Google account files photos into Drive by date/agent/lot, emails the agents when the last lot lands, and deletes runs after 13 days.

Auth & privacy

A team passcode is checked by the Google script, not in the browser, and entered once per device. Each lot is its own link, so a buyer who gets a forwarded email can never see someone else's home. Stage notes are neutral — no completion dates promised. Retention outlives an agent's forwarding delay.

Field-tested the hard way

Real failures came from stale copies of the app on phones and an old deployed script. So the app force-updates from a version file, and the backend has a test harness that runs Code.gs against fake Drive, mail, cache and locks through two full photo runs before anything is published.

04 MONSOON S13 · Hardware + firmware + simulator · Public

New wipers for a 1993 240SX

My friend Alex's 240SX needed its wiper and washer brain replaced. We designed a board that plugs into the factory stalk and the factory motor through a custom sub-harness — no wire in the car gets cut — and runs everything from one fused 12 V input. He measured on the bench; I did the schematic, layout, firmware and a test bench you can use in a browser.

3D render of the blue MONSOON S13 circuit board with three relays
MONSOON S13 v1.0. 120 × 85 mm, two layers, 2 oz copper, blue mask. Three relays, an ATtiny1616, and “Designed by Alex & Andrew” on the silkscreen.

The detail that decided everything

I read the scanned 1991–94 factory service manual page by page. The key finding: the wiper motor is switched on the ground side. The brushes run when they're grounded, and the auto-stop contact is grounded while sweeping and at +12 V when parked, which brakes the motor. Get that backwards and the whole output stage is wrong.

Brain
ATtiny1616, Arduino core, flashed over UPDI
Inputs
8 universal ADC channels read contacts and the delay ring alike
Modes
Park, intermittent 2–12 s (tapered ring), LO, HI, wash + fast-fast-slow + drip wipe
Safety
20 A fuse, TVS, reverse-polarity diode, park-fault timeout: 8 s with no auto-stop change → all off
Loads
Designed for 3–5 A running, 12–15 A stall (Alex's measurements)
Checks
Schematic generated by script, ERC/DRC clean, autorouted with Freerouting

The virtual test bench

It's built from the circuit board file itself — the netlist is solved in the browser and 18 design checks run on load, so mistakes show up before a board is ever ordered. Move the stalk, twist the ring, crank the engine.

Open the test bench here
05 Blockchain Surfer · DeFi scanner & executor · Private, retired Oct 2026

Six scanners, one rule: prove it first

A 24/7 service on the same Raspberry Pi that watched public DeFi data on Base, Arbitrum and Polygon every 15 seconds, scored what it found from 0 to 100, and only acted when a rules engine and an on-chain dry run both agreed. I shut it down and moved the Pi to home duty in October 2026.

The interesting part isn't trading. It's building something that handles real money and has to refuse far more often than it acts.

Mission Control
Mission Control, the Next.js dashboard, live over WebSockets: opportunity radar, the “block miner” that digs one layer per scan, the scoring core and an equity curve. Recorded with a demo data feed — the real wallet isn't connected anymore.
01 · Scan

Prices, yields, vaults, cross-DEX spreads, liquidation health factors, gas — six scanners concurrently every 15 s (FastAPI, asyncio, APScheduler).

02 · Score

0–100 from profitability, liquidity, confidence and competition, minus a volatility penalty. A small ML model ran in shadow mode first and earned weight only after 30, then 100 real outcomes.

03 · Rules

Minimum confidence 80, profit ≥ 3× costs, max 10% exposure per trade, daily loss limit, a review mode after consecutive losses, gas-spike and liquidity guards.

04 · Dry run, then act

Every transaction is simulated with eth_call first — a doomed one costs zero gas. The contract dry run is ground truth, not the scanner's math.

An atomic contract with one exit

I wrote and deployed a Solidity contract (Foundry, fork-tested 11/11) for flash-loan liquidations: borrow the debt from Aave, liquidate, swap the collateral, repay the loan plus 0.05%, keep the difference — all in one transaction. Every path ends in a single check: require(finalUSDC >= input + minProfit). If anything is off, the whole thing reverts. Worst case is the gas fee. The contract holds no funds at rest.

  1. “No opportunities” was a bug, not a quiet market.

    A chain audit found 30 liquidations in 14 days; we'd caught 0. The scanner only looked back 2.3 days for borrowers. A resumable backfill to the beginning of the chain plus risk-tiered health checks fixed it.

  2. docker compose restart doesn't reload env vars.

    A safety feature sat switched off for hours because the container kept its old environment. Recreate, then verify inside the running process.

06 Property OS · Full-stack SaaS · Private, in development, not launched

One system for a real-estate office

The same agents handle resale, new construction and new leads, in five different tools. Property OS puts lead intake and AI texting, construction lot tracking with photo runs, and the deal pipeline in one place, built to be sold per office. It's feature-complete as an MVP and private — no customers, no revenue, and I'll say exactly that.

Property OS · Overview
Property OS overview dashboard
Property OS · Messages · AI engine
Property OS AI messaging
Property OS · Leads
Property OS leads pipeline
Property OS · Onboarding
Property OS onboarding wizard

Screenshots from my dev machine with the sample data that ships in the repo.

Auth

Supabase Auth with @supabase/ssr session cookies in Next.js 16. The FastAPI side verifies the JWT on every request through a get_current_user dependency on all ten routers. The only public routes are the ones that must be: Twilio and Facebook lead webhooks and the Google OAuth callback.

Multi-tenancy

An orgs table and an org_id on all 15 data tables. Middleware reads the org from the JWT into a per-request ContextVar, and every service filters reads and stamps writes with it — one office can never see another's leads.

APIs

Twilio SMS with an AI agent whose tone and hours are configurable, Google Calendar booking, Stripe Checkout + Customer Portal + webhooks (returns safe mock data when keys aren't set, so dev never touches real billing), CSV and webhook lead intake.

07 The garage · Car projects · Private

I wrench on cars. Some of it is code.

Two builds for my red Model 3 Performance: a dashboard for the car's live data through Tesla's own API, and a light strip in the dash that greets me when the car wakes up and says goodbye after it shuts off — without cutting a single wire.

TeslaDash

localhost:3737
Running against a sample data feed for this page, so it never wakes the real car.
How the login works
  1. Register as a Tesla Fleet API partner.

    Tesla requires an app's public key at /.well-known/appspecific/com.tesla.3p.public-key.pem on a real domain. I served it through an ngrok tunnel just for the registration.

  2. OAuth with scopes.

    Sign in on auth.tesla.com; the callback swaps the code for an access token and a refresh token.

  3. Tokens stay on the server.

    Kept in a server-side file and refreshed before expiry; the browser never sees one. The server polls every 10 s and pushes over a WebSocket.

Dash LED: the wake-up and the goodbye

Wake → blind-spot warning (left) → goodbye, recorded from the simulator I wrote to design the animations before installing anything.

Wake

The ESP32 booting is the signal: the outlet gets power when the car wakes, so no signal wire is tapped. 90 LEDs split at the center and mirror outward.

Goodbye

A 10 kΩ / 3.9 kΩ divider turns 12 V into 3.37 V for the ADC. When it drops, two 10 F supercapacitors keep the ESP32 alive for the 3.8 s goodbye, then deep sleep.

Blind spot

The 2023 Highland has a proprietary diagnostic port and unpublished CAN IDs, so phototransistors read the factory A-pillar indicators optically instead.

Firmware

Custom-compiled WLED with my own usermod for the triggers, debounce and hold times; presets for wake, goodbye, ambient and each blind-spot side.

Play with the simulator

The actual design tool, embedded. Click presets, change speed and brightness.

Load the LED simulator
08 AI MT5 bot · Forex automation · Private

Read the log before you change the code

My forex bot had a losing stretch. Instead of tweaking indicators, I went through the closed-trade log line by line. The strategy wasn't the main problem — the risk setup was. Three structural causes:

  • Three positions, one bet. Up to three trades on the same pair in the same direction. When one was wrong, all three were.
  • 6% risk each, so a single bad signal could cost 18% of the account in a day.
  • The 5 PM wick. Repeated entries right before the session close, stopped out on the spike.
Worst single-signal loss
Open trades on one idea
6%
Setup
If the signal is wrong−18%

  • One position per symbol across EURUSD, GBPUSD and USDJPY; 2% risk; a hard 5% daily stop.
  • A session-close block plus a freeze around 5 PM; no off-hours entries.
  • M15 + H1 + H4 trend must agree; six entry strategies, each tagged so a bounded ML optimizer can tune its own thresholds every 30 minutes.
09 Build & play · Browser game · Unreal walkthrough

Designing the next house, too

The long-term plan is land and a barndominium. I modeled it in Fusion 360 and walk through it in Unreal Engine 5 with a PS5 controller — 45 feet wide, 10-foot walls, an 8/12 gable roof. And because planning a homestead is more fun as a game, I wrote one.

Play House Builder
What you can do

Pick 1–20 acres, draw walls corner to corner, generate a roof, paint, lay floors, dig a pond, plant beds, mulch, a greenhouse and string lights. Order supplies and a drone delivers them. Animals wander the lot.

Controls

Mouse + keyboard (B build, G paint, L floor, N pond, H greenhouse, I lights, Tab inventory) or a PS5 controller with a radial mode wheel. Click the game to capture the mouse; Esc to let go.

How

Plain JavaScript and three.js, ~11,000 lines, no build step. One module per tool: paint gun, path system, pond, greenhouse, drone.

10 Websites · Static, fast, hosted free on GitHub Pages

Sites for people I know

No frameworks, no build step, nothing to break. Push to main and it's live.

New England Pressure Cleaning
New England Pressure Cleaning website
New England Pressure Cleaning — client work for a real roof-soft-wash and pressure-washing business in Naples, FL, with sitemap and basic SEO.
Our Big Move
Our Big Move website
Our Big Move — our family's homestead vision, with a floor plan drawn to match a hand sketch exactly: every door where the red marks say.
Redesign concept
Finance firm website redesign concept
Redesign concept — an unsolicited pitch for a Naples finance firm. Labeled a proposal; not their live site.
11 Small tools · One-offs that solved one problem

Little things, done properly

Not everything needs to be a platform. These each fixed one real problem for me or someone I know.

drag to spin · 4,898 triangles

Charging dock, written as code

A desk dock for six USB ports and two device cradles, generated by a Python script against the Fusion 360 API: every pocket, wall and port cutout comes from measured dimensions, so changing a device means editing a number and re-running. A second script adds smooth topographic hills to the top face. Printed in PLA. This is the actual print file — spin it.

FL Land Hunter dashboard

FL Land Hunter

Tracks our search for 5–10 wooded acres in Southwest/Central Florida: flood Zone X, paved road frontage, under $100k. Instead of only watching listings, it links straight to each county's tax-deed auctions and code-enforcement records — the properties nobody's marketing. One HTML file; double-click to open; nothing installed.

Dashcam recovery

A neighbor's dashcam recorded a crash, but every file on the card “was corrupt.” Two separate causes: the recording was cut off by the impact, so the MP4's data-size field was literally 0 and players never found the index at the end; and the video was HEVC, which this Windows build can't decode. I patched the size field in copies, carved the newest footage by resyncing on HEVC frame boundaries, and transcoded to H.264.

The trap: the camera overwrites only the start of an old file, so the part that repairs cleanly is often the wrong day. I verified every clip against its internal timestamp and the time burned into the pixels — and flagged a date mismatch before it went to the insurer.

what the file said
00000018  00 00 00 00 6d 64 61 74  ....mdat
→ size 0: “runs to end of file”
…moov index sits near the end, never reached
patch: size = moov_start − 32
verify: creation_time ✓ · burned-in OSD ✓
ffmpeg: hevc → h264  plays ✓
How I work

Rules I actually follow

These come straight from the notes I keep at the top of every project.

It has to work the first time.

Plan it fully, explain it, get a yes, then build. No trial-and-error on a live system that a family depends on.

Test it where it lives.

On the real TV, in the real car browser, on the real phone in the field. Screenshots and frame rates, not “it worked on my machine.”

Never lock yourself out.

Back up before editing. Roll back automatically unless a change is confirmed. Know how to undo it before doing it.

Free first.

Most of what's on this page costs nothing to run: the Pi I already had, GitHub Pages, Google Apps Script, public data feeds.

Measure, then fix.

The bot's losses, the camera's 3.6-second lag, the missed liquidations — each was found by checking ground truth, not by guessing.

Write it so a busy person gets it.

Plain words, short sentences, outcome first. The house says “Someone's at the front door,” not “Person detected, zone 3.”

I build with AI coding tools — mostly Claude Code — the same way a shop uses power tools. I decide what gets built and how it should behave, I test it on the real device, I catch what's wrong, and I own the result. It's how one person ships hardware, firmware, back ends, front ends and documentation on seventeen projects.
About

Hi, I'm Andrew.

I started out with a soldering iron on aircraft lighting, ran my own automotive-electronics business for six years, and today I'm the last quality check on Teslas before they're handed to their owners. Fixing things and explaining them clearly to the person in front of me is the job I've always had. Software is how I do more of it.

I live in Southwest Florida with Jenna. I like cars more than is probably healthy.

  1. 2025 — now
    Delivery Vehicle Prep Lead Tesla

    The final quality gate before customer delivery: inspections, software verification, diagnostics, ADAS calibration support. Wrote the location's prep SOP and trained the team; site quality went from about 80% to a steady 95–100%.

  2. 2020 — 2025
    Technician & customer operations Kawi Performance

    Diagnosed complex mechanical and electrical systems, explained findings in plain language, and ran several customer projects at once from first call to final test.

  3. 2014 — 2020
    Founder & owner LED ATT

    Designed, built and installed custom automotive lighting. Ran the whole business solo — up to 100 customer conversations a day, Stripe operations, website, inventory — and grew an audience of 32,000+.

  4. 2010 — 2014
    Electronics & maintenance technician Aircraft Lighting International

    FAA-compliant electronics repair, soldering, testing and documentation, where mistakes carry real consequences. Prototypes from schematics, 3D-printed parts.

  5. Training
    Certified IPC Specialist, J-STD-001 Soldering

    Electrical engineering coursework at Farmingdale State College.