Platform · working name GIRDER TUBE · TRUSS · CLAMP · SIEVE

Turn the firehose
into evidence.

A platform that plugs into the customer's infrastructure and turns a regulated telemetry stream into four things at once — signed, error-bounded, queryable without decompression and complete. Each piece of the toolkit is one plane; the product is what stitches them together.

Today a customer gets, at most, one or two of these properties — never all four, and almost never with a certificate a regulator can run on its own.

Works with them — not instead of them

The regulated buyer won't ship raw data out, and won't replace the database.

Sidecar in the VPC or on-prem

Taps the stream from Kafka / broker / historian / ROS2 — it doesn't sit in the critical path. Their normal storage stays; the plane emits the signed archive alongside it.

The ε already lives in a document

TVE ≤ 1% in IEEE C37.118, half a cent in freight, sensor tolerance in the datasheet, recall floor in the SLA. We don't argue the number — it's already law or contract. Bring-your-own-ε.

The verifier is open

The guarantee can't be "trust us" — that would betray the very thesis: proof, not confidence. An open-source binary the regulator, the auditor or the counterparty runs on its own to check signatures and bounds.

The architecture

The five planes map 1:1 onto the pieces

One stream in; five guarantees out. What's measured against an installed incumbent comes with a number; the rest is a built primitive.

stream TUBE |error| ≤ ε signed Ed25519 · Merkle SIEVE search · zero FN CLAMP answer ±ε TRUSS watch · adversary open verifier regulator · auditor runs it alone
The same stream feeds all five guarantees; any third party re-runs the verifier to check.
01 · Ingest & bound measured

TUBE

Compresses with |error| ≤ ε guaranteed per sample, verified on decode. Circular channels (angle, heading); GOP index.

v4 ~30× · v5 ~53× deadband 85% CS2: SZ3 1.6–1.9× behind
02 · Seal primitive

signed / Merkle

Every block stamped with Ed25519 and committed to a Merkle tree. Tampering breaks the signature — it's the "tamper-evident archive" the rule requires, verifiable by a third party.

Ed25519 Merkle tree open verifier
03 · Answer measured

CLAMP

Certified aggregates ±ε over the compressed data, without decompressing. An interval that contains the truth: VWAP, compliance, climate envelope.

forecast 187/187 contained no decompression
04 · Search measured

SIEVE

Completeness: finds every occurrence of a pattern, zero false negatives. Certified range and k-NN, out-of-core.

hnswlib misses 9–22% 228× less I/O zero FN 5/5
05 · Watch measured

TRUSS + surprise

Counting and heavy-hitters robust under an adaptive adversary. Surprise as a byproduct of compression detects staleness and anomalies for free.

Redis counts 1 of 30,000 $2,999 evaded TRUSS blocks it
What the customer operates

One plane, three surfaces

  • Console — channels, bounds, surprise alerts.
  • SDK / tap — plugs into the broker, the historian or ROS2.
  • Verifier — handed to whoever needs to audit.
2026 · built and measured

The plan became operating infrastructure

On the same error contract, the pieces missing between the sensor and the audit — queue, black box, archive, joining, gaps, fleet, 2D perception and 3D space — twenty-three measured steps on real data (203 Hz IMU, ALOHA joints, car CAN, an AMI fleet, TUM depth video, KITTI LiDAR) with the experiment committed next to the code.

Transport measured

SLUICE — the queue

Under backpressure a normal queue can only drop — holes with undefined error. SLUICE widens the ε of the next blocks until the flow fits the link, then tightens back; the larger error travels declared in the record and is proved per record.

100% vs 41% at 0.4× the bandwidth ε declared and honored
Forensics measured

COFFER — the black box

Fixed flash that never forgets, only blurs: the past is re-compressed at larger ε with the blur on the label, the last seconds are untouchable, and incidents detected by multi-sensor voting stay pinned at full sharpness. On disk, crash-safe: a power cut doesn't lose the pinned incident.

24 KiB · 6 channels · incident at ε×1 reopens from disk after a crash
Retention measured

STRATA — the archive

Age-tiered retention without the downsampling lie: every query answers [lo, hi] guaranteed to contain the truth, by real timestamp (the nominal clock is off by dozens of periods — we measured), decoding only the window's boundary blocks.

6,000/6,000 windows contain the truth ~20 µs per query
Fusion measured

SPLICE — the join

Matching two sensors by nominal clock is a guess: on real CAN, nearest-timestamp gets 43.8% of pairs wrong in silence. SPLICE treats every timestamp as an interval and classifies each pair — CERTAIN, AMBIGUOUS or IMPOSSIBLE — never excluding the true one. The ruler from the measurement: certified fusion up to f ≈ 1/(4·ε_t).

nearest: 43.8% silently wrong 254,633/254,633 contained
Gaps measured

CAULK — the caulking

Signed physics (the datasheet slew) turns a data hole into a provable bracket — refuted with a number where it doesn't pay, and with unexpected gold: when physics says IMPOSSIBLE, the data contains a teleport. Dataset provenance auditing: the 39 violations in a public imitation-learning dataset were exactly its 39 episode splices.

39/39 splices · 0 FP 4:1 resampling proved
Fleet measured

TRESTLE — the federation

N edges upload only certified indices (~1 B/sample, 12× less than raw; the data never leaves the edge) and the cloud answers fleet aggregates with a guaranteed bracket. The step-1 lesson: changing the answer's contract (the EXACT total of a shifted, declared window) beat buying precision with bytes.

2,000/2,000 brackets adjusted window: 0.47%
2D perception measured

TUBE2D — the image

Images and depth video with per-pixel |error| ≤ ε, proved, plus certified region queries ("did the gripper zone stay within ±5 mm?") without decompressing. Beat SZ3 in all 4 cases — on video, ceding SZ3-3D the temporal axis and winning 1.24–1.33×. The measured lesson from the internal refutations: the axis that pays is the one the sensor correlates.

1.24–1.33× vs SZ3-3D verify 100% per pixel
3D space measured

PILE — the point cloud

LiDAR compressed with a per-coordinate bound and spatial absence proof: "no obstacle inside the safety zone" as a guaranteed verdict (EMPTY-PROVED / OCCUPIED-PROVED / UNDECIDABLE), zero false negatives by construction — the evidence ISO 26262/UL 4600 ask for, signable. On bytes, 0.8–2.8× of Draco, winning on dense fine clouds.

1,800/1,800 · zero FN dense fine: beats Draco
Multicast refuted

WEIR — the frozen one

Per-subscriber layered streaming: built, measured — and plain simulcast won on every metric. The tombstone is the first line of the README. The method on display: the same ruler that approves the three above rejects what doesn't pay.

simulcast won on every metric dead end documented

Bound never violated in any cell of any experiment. The technical section → · the plain-language version →

Where it fits

Which verticals — and the signal that separates a mandate from a nice-to-have

01

The ε already lives in a document

02

Buyer is compliance, not performance

03

Incumbent gives confidence, not proof

04

High-volume data, retained for years

05

A third party must verify

When 4–5 of these hold, the platform is an obligation, not a convenience.

Vertical ε comes from Planes that light up Status
AI platform
metering · RAG · training-data audit
SLA / quota / recall floor Watch · Search · Seal measured
Market surveillance
CAT · MiFID II / MAR
Rule 613, best-execution all five thesis · strong map
Games / replay-telemetry tick / perceptual Ingest · Answer measured
Synchrophasor / PMU
power grid
TVE ≤ 1% (IEEE C37.118) Ingest · Answer · Seal thesis
Robotics / ROS2
active lead (Instor)
sensor tolerance Ingest · Search · Seal stepwise plan
Cold chain · AMI · dams · EV battery health / safety / reg standard Ingest · Seal · Answer thesis
Freight
gwtFrete · 2B quotes
half a cent (step price) Ingest · Answer thesis
Meteorology / certified geo field tolerance Search · Answer measuredSZ3/ZFP home turf
Where to start

Anchor v1 on what's already measured

V1 · ANCHOR

AI platform

The only tier already measured against an installed incumbent (Redis 8.8, hnswlib). Acute, current pain; validates three planes without leaning on the "in-RAM loses" caveat.

TOP NARRATIVE

Market surveillance

The only place where the entire platform becomes a legal mandate. Use it as the top story — but treat it as thesis until a pilot with real tick data.

CHEAP PROOFS

Games & PMU

The two cheapest Ingest/Answer proofs to close next — one already measured (CS2 GOTV), the other with the ε written into the standard (TVE).

The honest limit

Where the requirement is just in-RAM speed, the platform loses.

Without completeness, without years of retention, without an adversary, without a third party verifying — an argsort or an in-RAM FAISS wins, and we've measured that (the imgemb lesson). The plane only pays off where the certificate is the deliverable. That's focus, not a bug: it turns down the wrong verticals.

A regulated stream that isn't proof yet?

Bring the ε that's already in your document — the standard, the contract, the SLA. We'll show which of the five planes lights up first and set up a pilot in your VPC.