Problem

Mineral exploration is expensive because geologists must determine the shape and location of an underground deposit before committing to drilling, with each drill hole potentially costing tens of thousands of dollars. Traditional computational models struggle to combine multimodal geological data, learn across multiple exploration sites, and make sophisticated inferences, leading companies to drill thousands of holes before a site becomes operational. Mineflow’s central use case is therefore helping geologists decide where and how much to drill, with the potential to reduce unnecessary holes and save millions in exploration costs.

Product / Service

Mineflow is an AI platform for mineral exploration. Geologists upload exploration-site datasets, after which Mineflow trains a custom deep-learning model that predicts mineral-deposit shapes and locations in two or three dimensions. Its earlier product positioning emphasized custom prospectivity modeling, while the current platform also turns exploration files and public datasets into searchable documents, map layers, drillhole views, 3D scenes, alerts, exports, and prediction workflows.

The product combines a large public exploration-data layer with customers’ own drilling, maps, and project files. Mineflow says its system automatically georeferences maps, links filings to projects, and lets users inspect drillholes, assays, geology, geophysics, terrain, and source documents in one workspace rather than through a separate GIS workflow. The site offers a free-start and demo path, while its pricing page describes a standard pay-as-you-go plan for early-stage teams.

Market

Mineflow competes in AI-powered mineral exploration and geoscience software, spanning SaaS, artificial intelligence and machine learning, and metals-and-mining applications. Its target users are mining companies and geologists working from initial exploration through advanced feasibility studies. Possible competitors identified in company databases include KoBold Metals, Terra AI, and Dassault Systèmes; KoBold and Terra AI are particularly relevant AI-native or AI-enabled comparables in subsurface and mineral exploration, although Mineflow’s emphasis is on a customer-facing exploration platform and custom deposit prediction.

Mineflow was founded in 2024 and entered Y Combinator’s Summer 2024 batch. It reported early trials with a lithium mining company in which its predictions were claimed to be more than an order of magnitude more accurate than competitors, and it has since opened access to a public dataset containing millions of drillholes and tens of thousands of airborne surveys, samples, faults, tenements, and mines. The company appears to be at an early commercial-validation stage: it has demo, free-start, and pay-as-you-go motions, and a third-party profile labels it as generating revenue, but that same profile does not disclose a current revenue figure or a paying-customer count. Reported funding is approximately $590,000, with Y Combinator and Karman Ventures among its investors.