Freelance and remote App developers jobs in Arusha, Tanzania

The Alliance of Bioversity International and CIAT is seeking a senior Android developer for a three-month consultancy to develop an in-app whole-plot image capture feature for the Ona Android application. Ona is a mobile and web digital phenotyping tool used by breeding teams in Sub-Saharan Africa. The application supports crop research for common bean, cowpea, sorghum and other crops, using images and machine-learning models to measure traits such as plant stand, flower count and pod count. The consultant will develop a reliable, production-ready feature that enables users to capture an image or image set of an approximately 4-metre plot without leaving the Ona app. The captured data must be automatically linked to the active plot and include the required metadata for automated plant-stand counting. The consultant will assess and prototype multiple approaches, including in-app panorama capture, guided multi-frame capture, short video sweeps, stitch-free counting and single wide-angle images. The selected approach will be integrated into the Ona app, tested on target Android devices and validated in the field with breeding teams. The work includes reviewing the existing Ona capture workflow and codebase, developing prototypes, evaluating image quality and device performance, implementing the selected solution, supporting field validation and handing over maintainable, documented code. Training new plant-stand models and redesigning unrelated Ona features are outside the scope of the consultancy. Requirements Technical Requirements Capture must take place entirely within the Ona app, without requiring users to switch to the native camera. The output must cover the full approximately 4-metre plot row at a resolution suitable for the plant-stand model. Capture and review should take no longer than the existing panorama workflow, with a target of less than 20 seconds per plot. The solution must work on the low- and mid-range Android phones used by Ona teams. Capture must work offline; any server-side processing must occur after synchronisation and must not block field use. The feature must operate in bright sunlight, wind and uneven terrain, with the phone handheld or mounted on the Bruno physical aid. Users must receive clear capture guidance, quality feedback and an easy option to retake unsuccessful images. Images must retain plot ID, timestamp, device information and capture settings within Ona’s existing data model. The implementation must follow Ona’s Android architecture and development standards and integrate with its existing backend and image pipeline. Third-party libraries must have licences suitable for use in a publicly funded, non-commercial tool. Required Qualifications and Experience At least five years of native Android development experience, particularly with Kotlin; Java experience is desirable. Experience shipping at least one application with a custom camera feature using CameraX or Camera2. Practical experience with image stitching, panorama development or computer vision, including tools or techniques such as OpenCV, feature matching, homography or image registration. Experience using phone sensors such as gyroscopes and accelerometers for capture guidance or motion tracking. Experience optimising applications for low- and mid-range Android devices, including memory, battery and heat management. Experience developing offline-first applications and background synchronisation. Fluent written and spoken English. Applicants must provide links to code, applications or demonstrations showing relevant camera, image-processing or stitching work. Desirable Experience Server-side image processing using Python, OpenCV or cloud functions. Video frame extraction. On-device machine learning using TensorFlow Lite, ML Kit or ONNX. Object tracking. Android NDK and C++ development. Experience in agriculture, field data collection or applications used in Sub-Saharan Africa. Experience contributing to open-source or publicly funded projects. Deliverables and Timeline Week 2 – D1: Inception note covering the current workflow, previous attempts, testing plan and device list. Week 4 – D2: Options report with working prototypes, test results and a recommended approach agreed with the Ona team. This is the go/no-go point. Week 9 – D3: Beta build integrated into the Ona app, including any required server component, ready for field testing. Week 11 – D4: Field validation report covering capture time, success rate and image quality for plant-stand counting. Weeks 12–13 – D5: Final release merged into the Ona codebase, with technical documentation, tests and a handover session. The consultancy is expected to last approximately 60 working days over three months. Fees will be paid upon acceptance of deliverables in three instalments: 30% for D1 and D2, 35% for D3, and 35% for D4 and D5. Working Arrangements Reporting: The consultant will report to the Head of Product and work closely with the Ona Android developers and machine-learning team. Location: Remote. Collaboration: Weekly progress calls, code reviews and pull requests through the Ona repository. Equipment and access: The Alliance will provide access to the codebase, relevant test devices or device lists, sample plot images and the Bruno aid where required. Intellectual property: All code, documentation and data produced will belong to the Alliance and may be released under an open licence. Data protection: The consultant must comply with Alliance data-protection and data-sharing policies. How to Apply Applicants should submit the following documents: A CV highlighting Android camera, image-stitching and computer-vision experience. Portfolio links, including applications, repositories or demonstration videos of similar work. A technical proposal. A financial proposal stating the daily rate and total fee for approximately 60 working days. Applications should be submitted through the Alliance jobs website at https://www.bioversityinternational.org/jobs/. The application must specify “Ona in-app plot imaging consultancy” as the position applied for. All application documents should be combined into one file named using the candidate’s last name and first name. Email applications will not be considered. Shortlisted applicants will participate in a technical interview with the Ona team, which may include a review of previous camera or image-stitching code. Additional Information Application deadline: 30 October 2026. Only shortlisted candidates will be contacted. The Alliance does not charge fees at any stage of the recruitment process. The Alliance will not request applicants’ bank-account information as part of the recruitment process. More information is available at https://alliancebioversityciat.org. About the Organization The Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT) develops research-based solutions that use agricultural biodiversity to transform food systems sustainably. Its work addresses malnutrition, climate change, biodiversity loss and environmental degradation. The Alliance is part of CGIAR, a global research partnership focused on achieving a food-secure future. Read more