For farmers
A clear path from need to a ready application.
Agri-Accept
Making every farmer loan-ready — turning farm, satellite, and weather data into bank-ready evidence.
Making every farmer loan-ready — turning farm, satellite, and weather data into bank-ready evidence.
Background
Agri-Accept began as Agri-Access, built by an HKUST undergraduate team for the APRU Tech Policy Hackathon (30 Sep – 2 Oct 2025) at Google Office, The Okura Prestige, Bangkok.
We placed third for transforming Earth observation into sustainable financial loans for Indonesia’s $25B credit gap — helping invisible smallholders become investable, while the bank keeps the decision.
Official coverage: APRU announcement · Finalist abstracts
The gap
Agriculture employs 30%+ of Indonesia’s workforce — yet 29 million smallholders remain economically invisible, with 72% below the poverty line (World Bank Group, 2020).
Roughly 20 million smallholders (~70%) cannot get formal loans; ~90% lack collateral, and about 20% turn to high-cost informal markets. KUR still asks for 3–5 years of paperwork most farms do not have.
Yield prediction alone cannot underwrite a loan — household shocks, prices, and timing break that shortcut. We separate agronomic risk, cash-flow capacity, and credit risk.
What we are
We turn farm, satellite, and weather data into bank-ready evidence. The bank keeps approve / review / reject.
A clear path from need to a ready application.
Independent farm proof and explainable risk for underwriting.
Shared readiness language and a file lenders can open.
Product flow
Match the right loan, close the gaps, and ship a verified evidence pack the lender can review.
KUR fit, gaps, and what to fix before apply.
Satellite + cash-flow proof, uncertainty-aware.
One-click passport — humans keep the decision.
How it works
Satellite, weather, soil, and cooperative records feed agronomic and cash-flow engines — then credit decision support under bank policy and human review.
Precedent
National and research schemes show EO-augmented underwriting is feasible — Indonesia’s rails are opening now.
Agri-fintech deployments have served 40,000+ smallholders with data-assisted credit (2025).
Research shows spatial variables improve credit-score accuracy beyond macroeconomic factors alone.
Banking use cases cut credit assessment cycles from ~15 days to a few days with alternative data.
Why now
OJK reforms encourage alternative credit scoring and weather-linked data. Open EO and AI weather (Sentinel, Earth-2) lower the cost of farm proof. KUR and AUTP provide subsidized credit and insurance rails — capital is flowing into agriculture.
Agri-Accept sits in that opening: data as collateral for farmers who lack assets, without replacing the lender.
Roadmap
Pilot with a West Java cooperative to refine explainable models and unit economics, then expand across provinces and integrate with state-owned banks supporting national KUR — targeting the $25B gap.
Live demo farms span rice, palm, coffee, cocoa, and dryland systems across the archipelago. After login, pick a farm and run Analyze Credit Risk on the private API.
Who we are
HKUST × Indonesia — product, capital, and engineering.
HKUST dual degree in Computer Science & Economics (T&M-DDP). Led Agri-Access to 3rd place at the APRU × Google Tech Policy Hackathon 2025.
LinkedInBSc Global China Studies, HKUST. Finance and strategy experience across MCP Asset Management, BCG, and CITIC Securities; co-founder of Fruitful.
LinkedInComputer Software Engineering, SMK Negeri 2 Kota Bekasi. Backend software engineer intern at PT. Traspac Makmur Sejahtera; FTC Nusantara Championship winner.
LinkedInLive pilot
Sign in to run analyses against the private Render backend. First call may take a minute while the service wakes. Request credentials if you do not have them yet.
Let’s make the invisible, investable.