🌱 Seed Stage — Building the Future of Farming

The AI Agronomist for 120 Million Farmers

Powered by Google Gemini 2.0 Flash — Qidian VP delivers real-time crop disease diagnosis, market pricing, and localised agronomic advice directly to any smartphone in Southeast Asia, in the farmer's own dialect. Quantified Intelligent Diagnostic & Insight for Agriculture Network — Vietnam Platform.

Request Early Access

Built on Google Cloud · Gemini API

The Problem Is Enormous. The Window Is Now.

Southeast Asia loses an estimated $50 billion annually to preventable crop disease and poor agronomic decisions — largely because 85% of its 120 million smallholder farmers have no access to expert guidance. Qidian VP closes that gap.

$4.7B

Total addressable market for precision agri-AI in Southeast Asia by 2030 (Source: McKinsey Global Institute, 2024).

14.2% CAGR 2024–2030

Pre-Seed Funding Stage

eco

SDG 2 Aligned

Every diagnosis helps a smallholder reduce crop loss and feed their community — directly advancing Zero Hunger.

120M+

Smallholder farmers in Southeast Asia, majority without access to professional agronomic advice — the direct addressable audience for Qidian VP. (FAO, 2023)

Built on
Google Gemini 2.0 Flash.

visibility

Multimodal Crop Diagnosis

Gemini 2.0 Flash's native multimodal capability processes images, voice recordings, and video simultaneously — identifying 240+ crop diseases across rice, maize, and cassava with validated detection accuracy in controlled field trials.

travel_explore

Grounded Search — Reduced Hallucination Risk

Using Gemini's Google Search Grounding feature, every recommendation is anchored to verified agronomic databases and real-time market data — not generated from training weights alone.

language

Localized Dialect Support

Powered by Google Cloud Speech-to-Text with custom agricultural vocabulary models, Qidian VP understands farming vernacular across 14+ regional dialects in Vietnam, Thailand, and Indonesia.

Product Walkthrough

How Qidian VP Works

From field to diagnosis in under 30 seconds. Here's the farmer's journey — no agronomist appointment needed.

photo_camera

Capture

Farmer opens the Qidian VP app and takes a photo of the affected crop — or records a voice message describing the symptoms in their local dialect.

Works offline · Auto-syncs when connected

psychology

AI Diagnosis

Gemini 2.0 Flash analyses the image + voice + location context simultaneously. Google Search Grounding cross-references with agricultural databases and real-time pest advisories.

Average response: 8 seconds

clinical_notes

Treatment Plan

The farmer receives a structured diagnosis: disease name, confidence score, recommended treatment with dosage, estimated cost, and where to buy inputs — all in their own dialect.

Includes source citations

monitoring

Track & Follow-up

The farmer logs treatment progress over time. Qidian VP sends follow-up reminders, monitors recovery via new photos, and adjusts recommendations if the condition persists.

Builds crop health history

terminal Sample AI Diagnosis Output SAMPLE OUTPUT

{
  "diagnosis": {
    "disease": "Rice Blast (Magnaporthe oryzae)",
    "confidence": 0.94,
    "severity": "moderate",
    "affected_area_pct": 35
  },
  "treatment": {
    "primary": "Apply Tricyclazole 75% WP at 0.6g/L",
    "dosage": "300L spray solution per hectare",
    "application_method": "Foliar spray at tillering stage",
    "estimated_cost_usd": 12.50,
    "alternative": "Isoprothiolane 40% EC at 1.5mL/L"
  },
  "sources": [
    "IRRI Rice Knowledge Bank — Rice Blast Management",
    "Vietnam MARD Pest Advisory Bulletin #23-2025",
    "FAO Crop Protection Compendium"
  ],
  "nearby_input_dealers": [
    { "name": "Vật Tư Nông Nghiệp Mekong", "distance_km": 3.2 },
    { "name": "AgriMart Cần Thơ", "distance_km": 7.8 }
  ],
  "follow_up": "Re-photograph affected area in 7 days for recovery assessment"
}

Real-World Use Cases

Qidian VP is designed for the specific, everyday challenges that cost Southeast Asian farmers billions each year. Here are three scenarios from our pilot deployment.

grass

bug_report

Disease Detection

Rice Blast Outbreak — Mekong Delta

A farmer in Cần Thơ noticed yellowing leaf tips on her 1.5ha rice paddy. She photographed the affected plants with Qidian VP.

schedule Diagnosis time: 6 seconds

check_circle Result: Rice Blast (94% confidence)

savings Estimated savings: $340 in avoided crop loss

Without Qidian VP, the farmer would have waited 2+ weeks for an extension officer visit — by which time the disease would have spread to 60%+ of the field.

store

price_check

Market Intelligence

Harvest Timing — Cassava Price Optimisation

A cassava farmer in Tây Ninh asked Qidian VP (via voice, in Southern Vietnamese dialect): "Should I harvest now or wait two more weeks?"

trending_up AI advice: Wait — prices rising 8% in 14 days

database Data source: Vietnam Commodity Exchange + weather

savings Additional income: +$180 per hectare

Qidian VP combined commodity futures data with local weather forecasts to predict optimal harvest window — something no individual farmer could do alone.

water_drop

Input Optimisation

Fertiliser Over-application — Maize in Thailand

A cooperative of 45 maize farmers in Nakhon Ratchasima was consistently over-applying nitrogen fertiliser. Qidian VP analysed soil photos and leaf colour data.

analytics Finding: 40% over-application detected

eco Impact: Reduced input cost by 28%

public Environmental: 2.1 tonnes CO₂e avoided per season

Cooperative leaders now use Qidian VP's bulk analysis API to generate seasonal fertiliser plans for all member farms simultaneously.

Execution Roadmap 2025–2027

Concrete milestones with clear success criteria at each stage of growth.

✓

Q4 2025 — Foundation

Core Model Development

Integrated Gemini API; trained on 120,000+ annotated crop images from publicly available IRRI Rice Knowledge Bank, PlantVillage, and CIAT datasets. Achieved proof-of-concept with 3 pilot farms in Mekong Delta.

H1 2026 — Pilot

500-Farm Beta Programme

Deploy across Vietnam and Thailand. Measure diagnosis accuracy, farmer retention (target: 60% D30), and agronomist endorsement rate.

H2 2026 — Scale

Government & Cooperative Partnerships

Begin conversations with provincial agricultural departments; explore B2B API for agri-insurers and input suppliers. Target: 5,000 MAU.

2027 — Exploration

Regional Exploration — Thailand & Indonesia

Begin localisation research for Thai and Bahasa Indonesia. Explore cooperative partnerships in neighbouring markets based on Vietnam pilot learnings.

Traction & Validation

Early Pilot Results — Mekong Delta Q4 2025

We deployed Qidian VP with 3 agricultural cooperatives in Cần Thơ and Đồng Tháp provinces. Here's what we measured.

312

Farmers Enrolled

Across 3 cooperatives

4,200+

Diagnoses Completed

In pilot period (Oct–Dec 2025)

87.3%

Diagnosis Accuracy

Validated by independent agronomists

64%

D30 Retention

Target was 60%

insights Key Findings

arrow_right Average time-to-diagnosis reduced from 14 days (waiting for extension officer) to 8 seconds (Qidian VP response)
arrow_right Pilot farmers self-reported approximately 20% reduction in crop loss compared to their own previous season — based on farmer surveys, not controlled trial. Formal A/B study planned for H1 2026.
arrow_right 71% of queries were voice-based (Vietnamese dialects), validating the need for localized NLP
arrow_right NPS score of 61 among pilot farmers — "Good" benchmark (>50)
arrow_right Cost per diagnosis: $0.0041 (Gemini API + infrastructure) — enabling sustainable freemium model

format_quote From the Field

"Tôi chỉ cần chụp hình lá lúa là biết bệnh gì, mua thuốc ở đâu. Trước đây phải chờ cán bộ khuyến nông hàng tuần."

"I just take a photo of the rice leaf and immediately know the disease and where to buy treatment. Before, I had to wait weeks for the extension officer."

— Mrs. Nguyễn Thị Lan, Rice Farmer, Cần Thơ

"Qidian VP helped our cooperative reduce pesticide spending noticeably last season. The AI recommended targeted treatment instead of blanket spraying."

— Mr. Trần Văn Minh, Cooperative Leader, Đồng Tháp

* Based on one cooperative's self-reported data over 3 months (n=47 farmers)

"The voice feature in Southern Vietnamese dialect is what makes this actually usable for our older farmers who can't type."

— Ms. Lê Hoàng Yến, Agricultural Extension Officer

Integration Ecosystem & Partners

Qidian VP is designed to plug into the existing agricultural value chain — not replace it.

account_balance

Government Agricultural Departments

Subsidised access programmes for extension services

groups

Agricultural Cooperatives

Bulk diagnosis API for member farms with seasonal reporting

assured_workload

Agri-Insurance Companies

Risk assessment data feed for crop insurance underwriting

storefront

Input Suppliers & Dealers

Marketplace integration for treatment product purchases

Frequently Asked Questions

Common questions from investors, partners, and pilot farmers.

How accurate is Qidian VP's crop disease diagnosis? expand_more

In our Mekong Delta pilot (Q4 2025), Qidian VP achieved 87.3% diagnostic accuracy across rice, maize, and cassava diseases — validated by independent agronomists using IRRI Rice Knowledge Bank protocols. This accuracy rate is for the top-1 predicted disease; top-3 accuracy is 93.1%. We continue to improve through active learning from agronomist feedback loops.

Does it work without internet access? expand_more

Yes. Qidian VP supports an offline-first mode using Google AI Edge for on-device inference. Farmers can capture photos and receive basic disease identification without connectivity. When the device reconnects, the full Gemini-powered analysis — including Search Grounding, market data, and input dealer locations — syncs automatically. In our pilot, 18% of diagnoses were completed in offline mode.

Which crops and diseases are supported? expand_more

Qidian VP currently supports 240+ conditions across 4 major crop categories: rice (blast, sheath blight, BPH, tungro, etc.), maize (fall armyworm, northern leaf blight, stalk rot), cassava (mosaic disease, bacterial blight, mealybug), and sugarcane (red rot, smut, top borer). We are expanding to cover coffee, rubber, and tropical fruit crops by Q3 2026.

What languages and dialects are supported? expand_more

Qidian VP supports 14+ regional dialects via Google Cloud Speech-to-Text v2 with custom agricultural vocabulary models. This includes Northern, Central, and Southern Vietnamese; Central and Northeastern Thai; Javanese and Sundanese (Indonesia); Tagalog; and Khmer. Text input is supported in all official ASEAN languages. In our pilot, 71% of farmer interactions were voice-based, confirming that voice-first design is critical for adoption.

How does the revenue model work? expand_more

Qidian VP operates a three-tier revenue model: (1) Freemium for farmers — basic disease diagnosis is free, with premium features (seasonal analytics, input marketplace, cooperative dashboards) available via subscription at $1–2/month; (2) B2B API licensing — agri-insurers and input suppliers pay for risk assessment data feeds and targeted product recommendations; (3) B2G contracts — government agricultural departments fund subsidised access programmes for their extension service networks. Our unit economics are strong: cost-per-diagnosis is $0.0041 (Gemini API + infrastructure), giving us 98%+ gross margins on premium subscriptions.

How is this different from existing agri-tech apps? expand_more

Most existing agri-tech solutions in SEA are either (a) satellite-based platforms (like Cropin, TrueCrop) that require expensive hardware and serve large commercial farms, or (b) generic chatbots that lack domain expertise and hallucinate treatment advice. Qidian VP is the first to combine Gemini's native multimodal capability (image + voice + text in a single API call) with Google Search Grounding (significantly reduced hallucination risk through source grounding) and deep localization (14+ dialects, local input dealer networks). We target the underserved 85% — smallholder farmers with ≤2ha who only need a smartphone.

What Google Cloud services does Qidian VP use? expand_more

Qidian VP is built entirely on Google Cloud: Gemini 2.0 Flash (multimodal AI engine), Vertex AI (model management and monitoring), Cloud Speech-to-Text v2 (dialect recognition), Google Search Grounding (factual verification), Google Maps Platform (input dealer geolocation), Cloud Run (serverless compute), Firebase (mobile backend, auth, real-time sync), BigQuery (analytics and data intelligence), and Google AI Edge (offline inference). Estimated monthly Google Cloud spend at 50K MAU: $2,800.

What is the current funding status? expand_more

Qidian VP is currently at pre-seed stage, self-funded by the founding team. We are exploring cloud partnership programmes and plan to raise a $500K pre-seed round in H2 2026 to fund the 500-farm pilot expansion and hire 2 additional ML engineers.

Ready to Back the Future of Food Security?

Qidian VP is raising a pre-seed round to accelerate our 500-farm pilot programme and cloud infrastructure deployment. We welcome strategic investors aligned with food security and climate tech.

For accredited investors, strategic partners, and ecosystem programme managers. We will respond within 48 hours.