Founder

Tyler Poniatowski builds Fleet Street GPS himself — hardware, firmware, and all.

Not a team of ten with a hardware vendor and a dashboard contractor. One person hand-soldering trackers, writing the firmware that runs on them, and running the backend they report to — because a shipment-tracking product you don't build yourself is one you can't actually vouch for.

Why this exists

The problem isn't that freight moves. It's what happens the moment it's out of sight: a shipment goes quiet, and the first anyone hears about it is a customer asking where it is.

Fleet Street GPS started from that gap — the space between "we shipped it" and "it arrived," where a lot of operators are flying blind. Carrier scan updates are hours apart at best. Consumer trackers like AirTags need a nearby phone to report anything. Neither one tells you the moment something's actually wrong.

The fix isn't a smarter dashboard bolted onto someone else's hardware. It's a tracker built to report continuously on its own, log everything even when the network doesn't cooperate, and flag the exact moment it goes silent — built and tested by the same person who has to answer for it when it doesn't work.

Background

Over a decade in identity and access management and IT security, building startups through rapid growth and solving the kind of high-stakes problems where the cost of getting it wrong is measured in profitability, not just inconvenience.

That background shows up directly in how Fleet Street GPS is built, not just in the pitch. Tiered admin/ops roles, a full audit log of every account and device change, and access scoped so a customer only ever sees their own shipments — that's IAM discipline applied to a fleet dashboard, not a feature added because a checklist asked for it.

The same instinct that goes into securing who can touch what goes into the tracking itself: a system built around the failure modes that actually cost money — a shipment going dark, an outage nobody catches until a customer complains, a chain of custody nobody can prove after the fact — because that's the kind of problem this experience was built to solve.

Built and proven firsthand, not outsourced

Three things that are true about how this got built, not marketing copy about how it should work.

🔧

Hand-built hardware

Every tracker starts as a Seeed XIAO ESP32S3 board, hand-soldered and flashed personally — including a browser-based flashing tool built specifically so new units don't need a laptop with PlatformIO installed to bring online.

🚗

Tested on a real route, not a bench

Before geofence arrival alerts shipped, they were proven on an actual multi-hour drive — departing at 6am to validate live tracking and arrival detection under real driving conditions, not a simulated GPS feed.

🩺

Outages get diagnosed, not shrugged off

When a tracker went dark for over 36 hours, the fix wasn't a canned response — it was tracing the failure live over a serial monitor down to a specific SD card fault, then shipping a fallback so a failed card degrades gracefully instead of going silent.

A tracker that only works when everything goes right isn't a tracking product — it's a demo. The whole point is knowing the moment something doesn't.

What that means day to day

The same three things every feature gets built and tested against.

Clarity

Status at a glance

No guessing whether a shipment is healthy, delayed, or dark — it's one look at the fleet map.

Speed

Alert while it's still fixable

The value is catching a problem in time to act on it, not documenting it after the fact.

Continuity

The trail survives the gaps

Coverage isn't perfect. Store-and-forward logging means a dead zone doesn't erase the route.

See it running before you commit to anything

Reserve early access — no payment, founding-customer terms, cancel anytime.

Reserve yours