Precision livestock intelligence

Detect disease 4–7 days early in every pig

RFID + IR checkpoints monitor 10 health parameters per pig — temperature, feeding behavior, circadian rhythm, activity — building individualized baselines that catch illness before humans can see it.

Request demo How it works

Live system schematic

Scroll

01 — The problem

The US pork industry loses $1.2B/year to disease it finds too late

Finishing pigs — 73 million head, the largest unmonitored segment — have zero funded individual-level health monitoring. Illness is detected by eye, days after intervention could have started.

Annual PRRS losses — US

Source: National Pork Board

Finishing mortality rate

Industry average

Loss per barn per year

Preventable with early detection

02 — The solution

Individual monitoring at $1.75/pig/year

Passive ear tags. Barn-mounted checkpoints. AI that learns each pig's personal baseline and flags the ones going wrong — days before a human would notice.

Individual identification

Passive UHF RFID ear tags identify every pig at every checkpoint. No batteries, no charging, no manual scanning.

Surface temperature

IR sensors at feeder checkpoints capture body temperature. Fever detection without restraint, stress, or contact.

Behavioral analytics

Activity level, feeding frequency, circadian rhythm, zone dwell — all computed from checkpoint data, per pig, per day.

Individualized baselines

Rolling 7-day baselines per pig. Every animal compared to itself — catching what herd-level monitoring misses.

ML anomaly detection

Multi-parameter scoring with human-readable reason codes. "Pig #247: feeding -40%, temp +1.2°F" — not a black-box score.

Offline-first edge

Raspberry Pi runs the full engine with no internet. Data syncs when connectivity returns. Zero downtime in the barn.

03 — How it works

From barn sensors to actionable alerts

Passive hardware. Zero farmer workflow change. Pigs walk past checkpoints on their own schedule.

Tag & install

Passive ear tags. Antenna + IR per pen feeder. Pi in the barn office. Done in a day.

Baseline

Every feeder visit logs identity, temperature, timing. 7-day rolling baselines build per pig automatically.

Detect

ML engine scores each pig across 10 parameters vs. its own baseline. Anomalies get plain-English reason codes.

Act

Dashboard shows vet worklist. Most critical pigs first. Daily digest at 6 AM. Resolve alerts in one tap.

03b — See it live

What the farmer actually sees

From barn floor to dashboard alert — a walkthrough of the full product experience.

Auto-cycling

Pen A — 24 pigs

#201

#202

#203

#204

#205

#206

#207

#208

#209

#210

#211

#212

RFID+IR checkpoint — idle

Live readings

Pig—

Temp—

Feeding—

Activity—

Score—

Status—

247

Healthy

Flagged

Critical

98.4%

Sync rate

Pig	Score	Temp	Feeding	Status
#247	23	104.2°F	−42%	Critical
#183	41	103.1°F	−28%	Warning
#092	38	101.8°F	−31%	Warning
#156	82	101.2°F	+3%	Healthy
#301	91	101.0°F	+1%	Healthy

Critical · Score 23

Today 5:42 AM

Pig #247

Feeding frequency dropped 42% below 7-day baseline. Surface temperature elevated 2.8°F above individual norm. Activity declined 31% over 2 consecutive days. Circadian rhythm irregular since yesterday.

Temperature

104.2°F (+2.8°F)

Feeding

−42% vs baseline

Activity

−31% (2 days)

Circadian

Irregular

Disease pattern match

Best match: PRRS

PRRS

78%

Influenza

45%

PED

12%

Recommended actions

Isolate immediately

Move #247 to isolation pen. Prevent contact with pen-mates. Mark pen for observation.

Request vet review

Pattern suggests PRRS. Recommend PCR test to confirm. Collect blood sample for lab submission.

Monitor pen-mates

11 pigs in Pen A share feeder with #247. System will auto-flag any pen-mate deviations within 24h.

Log treatment

Record treatment type, dosage, and date in the resolve modal. Outcome data improves future detection accuracy.

This is what the vet sees at 6 AM — before walking into the barn.

04 — Traction

Pilot confirmed. Industry engaged.

Built the product. Secured the pilot. Earning trust from the largest names in US pork.

TTU pilot farm — confirmed

New Deal Swine Facility, Texas Tech University. Hardware install on funding close.

Smithfield Foods — interest expressed

Contact within Smithfield expressed product interest. Follow-up active.

Seaboard Foods — C-suite intro

Top-10 US pork producer. C-suite contacts via advisor network.

15,000-pig commercial farm

Warm introduction. Active interest in commercial pilot post-TTU validation.

National Pork Board chief vet

Introduction in progress via TTU advisor network.

Full platform — deployed

Edge → Cloud → Dashboard → Workdesk. Live on Railway + Vercel. Customer provisioning operational.

USDA ARS

Funk et al. (2024): 2,826 grow-finish pigs, 2.8-day early detection from feeding behavior alone. InscaleBio adds 9 additional parameters. That's the validated floor — not the ceiling.

05 — Team

Built by builders

Ishaan Das

Co-founder · Technical

B.S. Mechanical Engineering, Texas Tech University. Built the edge compute system, ML detection engine, and full-stack platform from scratch. RFID/sensor hardware, anomaly detection, prototype engineering.

HardwareMLFull-stackEdge compute

David Graziano

Co-founder · Business

B.S. Agricultural Economics, Texas Tech University. Customer discovery, integrator relationships, pilot farm recruitment. Leading Smithfield, Seaboard, and National Pork Board outreach.

SalesAg economicsBDOperations

Advisor

Siddhant Das — IIT Bombay EE/AI · Qure.ai