Hive Mind is an application for beekeepers to manage apiaries, view terrain and forage analyses on a map, check the weather, plan events, save notes, and use the Maya AI assistant.
You can create an account or log in using email, Google, or Apple. After registration, you may need to accept the terms and conditions and verify your email address before entering the app.
The Dashboard shows a welcome message, an overview of apiaries, the latest notifications, and upcoming events.
Go to My apiaries, tap Add apiary, enter the apiary name, select its location on the map, and optionally add an address, number of hives, and description. The name and location are required.
Your current plan may limit the number of apiaries you can create. You can check or change your plan in the Subscription or Plans sections.
Apiary details include the number of hives, apiary rating, detailed rating breakdown, land cover analysis, forage analysis, weather, assigned events, and related assistant sessions.
Open the Map from the bottom navigation menu. You can search for places or apiaries, view active and inactive apiaries, show your location, use measurement tools, toggle weather layers, and run analyses.
The app supports terrain analysis and forage analysis. For forage analysis, select a point on the map and the month and year of satellite imagery, then press Analyze.
Saved analyses are available in the Analyses Archive, where you can filter them by year, type, date range, and sort order. You can also add analyses to favorites or delete them.
Go to Events, tap Add event, select the date and time, enter a title, optionally assign the event to an apiary, add a description, and set reminders.
Go to Notes, tap Record note, grant microphone access, record the audio, then add a title, optionally an apiary and a category before saving. The maximum recording length is 10 minutes.
Maya is a built-in AI assistant. You can select an apiary, start a voice session, talk to the assistant, and then end the session to generate and save a summary or note. Sessions can last up to 15 minutes.
Notifications appear in the Notifications tab and may also be visible on the Dashboard. They can lead to related app sections, such as analysis results or apiary details.
Open My profile to edit personal and beekeeping details, change the language, manage analytics consent, change your password (if available), log out, or delete your account.
Use the Subscription section to view your current plan, and Plans to compare available subscription options.
Hive Mind helps increase efficiency primarily through better apiary location selection - the beekeeper places hives where there is a real chance for the most and most stable forage, rather than "by feel". As a result, bee colonies work in more intensely nectar-yielding landscapes, which, according to pollination research, translates into a several to even several dozen percent increase in crop yields and potentially 20–30% more honey per season.
Data on forage directly translates into business decisions. It shows where the bees' work will be most economically beneficial. A beekeeper can:
The difference is fundamental. Google Maps are often images from a few years ago that are used for field orientation. Hive Mind provides dynamic data (updated weekly). We don't just show the image, but analyze it - our algorithms see what flowering phase a given plant is in and whether its nectar potential is growing or falling.
No. Our goal was to create a tool that "translates" complex satellite data into a simple language of benefits. The beekeeper doesn't see raw radar readings but an intuitive map with colorful potential zones. Using the app is as simple as using popular map apps on your phone.
It is not required. Hive Mind works based on remote sensing (remote data), so you can use analyses without having a single sensor in the hive. However, if you have hive scales, the data from Hive Mind is a perfect complement to them - the scale will tell you that honey is increasing, and our app will explain where the bees are bringing it from and where it is worth placing more colonies.
The app is designed for field work. Key maps and forage data can be downloaded to the device's memory (offline mode), allowing for efficient activity planning even deep in the forest or on remote wastelands where cellular network coverage is limited.
Yes, we are working on this as a core feature. Our AI models are learning to recognize the spectral signatures of specific forage. Thanks to this, the user can not only check "if it's green", but receive information about the probability of occurrence of specific crops or clusters of melliferous trees in a given region.
Yes, the app includes a migration recommendation module. The algorithm assesses the forage potential in time and space, and then can send the beekeeper a notification that moving hives to another zone in a specific time window makes business sense (e.g., due to the approaching peak of a specific forage).
Currently, the system operates at the level of general forage zones, not individual plots. The entire space is divided into a hexagonal grid, and for each hex, the probability of the occurrence of valuable forage is calculated - today we use hexes with a side of about 50 m, and in the forage layer, zones with a radius of about 2 km are analyzed. Ultimately, we want to move towards indicating specific areas (separated patches of vegetation) along with their estimated forage volume and clear boundaries in the field.
The entire core of the analysis is based on machine learning methods, rather than classic, manually set remote sensing indices. We build neural networks that learn to recognize characteristic patterns of vegetation and changes over time visible in satellite images, and traditional remote sensing methods serve mainly for preliminary data preparation, quality control, and validation of results.
Erend Space combines radar (SAR) and optical satellite data. Both types of data have a spatial resolution of 10 meters, which allows analyzing the landscape structure at a level useful for forage planning.
We use data with a resolution of about 10 meters and a refresh rate of 7 days. For a beekeeper, this means seeing landscape structures at the level of larger fields, meadows, and vegetation complexes, rather than individual trees or bushes - accurate enough to assess the forage potential within a few kilometers of the apiary.
Satellite data is updated on a weekly cycle - a new image appears roughly every 7 days. Models learn on a three-month window: from each month we choose the best few photos (a total of about 9 scenes for one sample) to capture vegetation dynamics and minimize the impact of random noise.
To limit the impact of clouds and vegetation stagnation, we work on a wide time window covering at least three months of the growing season. From each month we select only those images where cloud cover is below the threshold (20%), and the photos themselves come from the spring-summer period, when vegetation provides the most useful signal for the models.
We base our backend infrastructure on production-standard cloud solutions, specifically the Firebase platform. User data - apiary locations, production information, plot data - is handled in accordance with GDPR requirements, and security is supported by authentication, authorization, and transmission security mechanisms offered by the cloud provider, as well as our internal procedures.
To develop similar solutions in other industries, we first need a stable prototype that actually works in beekeeping and confirms the effectiveness of the approach. Natural development directions include:
