[{"content":"The SubAlpine level of a mountain lies between the summit and the foot of the mountain, where, metaphorically speaking, ideology and reality are always taken into account in the realization of a project. An idea can only be realized if it is designed to be stable, relevant and of common interest.\nThis is why SubAlpine.IT offers services in response to projects that generate interest, where new machine learning methods are carefully selected, without exceeding the dimensions of the need. In a new era, resilience and habit change must be at the heart of every project.\nValues # Proportionality — the simplest method that solves the problem is the right one. A well-calibrated statistical model often beats an oversized neural network, in cost, robustness and maintainability alike. Sobriety — sensitive to current ecological and climate issues, SubAlpine.IT pays attention to the computational and energy footprint of the solutions it designs. Rigour — an academic research background guarantees methodological soundness: measurable objectives, validated results, reproducible experiments. Autonomy — every mission ends with a transfer of knowledge, so that the solution remains usable and maintainable without dependency. Practical information # SubAlpine.IT is based in France and operates remotely or on site, for missions of any size — from a few days of consulting to long-term R\u0026amp;D collaborations.\n📫 Reach out at contact@subalpine.it — or read more about the founder on the Story page.\n","externalUrl":null,"permalink":"/en/about/","section":"About","summary":"","title":"About","type":"about"},{"content":"","externalUrl":null,"permalink":"/en/authors/","section":"Authors","summary":"","title":"Authors","type":"authors"},{"content":"","externalUrl":null,"permalink":"/en/categories/","section":"Categories","summary":"","title":"Categories","type":"categories"},{"content":" Site publisher # The subalpine.it website is published by:\nJérôme Buisine — sole trader (micro-entreprise) operating under the trade name SubAlpine.IT\nSIREN: 822 432 902 — SIRET (head office): 822 432 902 00044\nActivity: computer programming\nRegistered office: Appartement E102, Bâtiment E, 129 impasse des Embrunes, 74230 Dingy-Saint-Clair, France\nContact: contact@subalpine.it\nPublication director: Jérôme Buisine.\nHosting # This website is hosted by:\nOVH SAS\n2 rue Kellermann, 59100 Roubaix, France\nPhone: +33 9 72 10 10 07 — www.ovhcloud.com\nIntellectual property # All content on this site (texts, logos, illustrations, layout) is the property of Jérôme Buisine unless stated otherwise. Any reproduction, representation or distribution, in whole or in part, without prior written authorisation is prohibited.\nLogos and trademarks of the third-party projects mentioned (StratAlp, Swoop, Meristem) remain the property of their respective owners. This site is powered by Hugo and the Blowfish theme, distributed under their respective licences.\nPersonal data # This site sets no tracking cookies and uses no analytics tool. A display preference (light/dark theme) may be stored locally in your browser (localStorage); it is never sent to any server.\nThe only personal data processed is what you voluntarily send by e-mail to contact@subalpine.it. It is used solely to answer your request and is neither shared nor used for commercial purposes. In accordance with the GDPR, you have the right to access, rectify and erase your data, which you may exercise at the same address.\nWeb fonts are loaded from Google Fonts; your IP address is therefore transmitted to Google\u0026rsquo;s servers when the page loads. See the Google Fonts privacy FAQ for details.\nLiability # The publisher strives to keep the information on this site accurate and up to date, but cannot guarantee the absence of errors or omissions. External links are provided for information only; the publisher is not responsible for the content of third-party sites.\n","externalUrl":null,"permalink":"/en/legal/","section":"Legal notice","summary":"","title":"Legal notice","type":"legal"},{"content":"Alongside client missions, SubAlpine.IT designs and operates its own products. They share the same philosophy: useful tools, built at the right scale, with a careful eye on data quality and resource consumption.\nStratAlp — running \u0026amp; trail strategy A race-strategy platform for runners and trail runners: build your pacing plan from the course profile, anticipate effort distribution and follow your progression. Mobile application backed by a dedicated API. Flutter \u0026middot; FastAPI \u0026middot; PostgreSQL stratalp.io \u0026rarr; Swoop — budget \u0026amp; wealth projection A budgeting and wealth-projection application: simulate your budget, project your savings and assets over time, and compare scenarios to make informed financial decisions. Flutter Web \u0026middot; FastAPI \u0026middot; PostgreSQL In development \u0026mdash; soon at swoop.subalpine.it Meristem — resilient tree planting, at the scale of France A mapping, assessment and action platform for tree planting: Meristem inventories and evaluates the quality of plantations, raises awareness, and connects places to improve with the right decision-makers and volunteers. An explainable, auditable geospatial rules engine — no black box. FastAPI \u0026middot; PostGIS \u0026middot; Next.js \u0026middot; MapLibre In development \u0026mdash; soon at meristem.subalpine.it ","externalUrl":null,"permalink":"/en/projects/","section":"Projects","summary":"","title":"Projects","type":"projects"},{"content":"","externalUrl":null,"permalink":"/en/series/","section":"Series","summary":"","title":"Series","type":"series"},{"content":"SubAlpine.IT supports companies, laboratories and public institutions in the design and implementation of data-driven solutions. Each mission starts from your actual need — not from a technology looking for a problem.\nMachine Learning \u0026amp; AI solutions # Design, development and deployment of machine learning models adapted to your use case:\nSupervised and unsupervised learning (classification, regression, clustering) Deep learning architectures for image, signal and structured data Reinforcement learning for sequential decision-making problems Model auditing, robustness evaluation and noise characterisation Data Science \u0026amp; statistical analysis # Turn your raw data into reliable, actionable knowledge:\nData exploration, cleaning and feature engineering Statistical modelling and hypothesis testing Dashboards and visualisations to support decision-making Reproducible analysis pipelines Optimisation # Reduce time, cost and resource consumption of your processes:\nMetaheuristics and local search methods for complex combinatorial problems Trade-off analysis between solution quality and computation budget Integration of optimisation loops into existing workflows Software development # Beyond data-centric work, SubAlpine.IT designs and builds complete software products:\nBackend development in Python (FastAPI), REST APIs and third-party integrations Relational and geospatial databases (PostgreSQL, PostGIS) Cross-platform mobile and web applications (Flutter) Packaging, deployment and operations (Docker, CI/CD, Nginx) Scientific consulting \u0026amp; R\u0026amp;D # Benefit from an academic research background applied to industrial problems:\nState-of-the-art reviews and feasibility studies Prototyping of novel methods and proof-of-concept development Scientific writing, publications and grant proposal support How a mission unfolds # Understand — a first exchange to frame the need, the data and the constraints. Assess — a short study to validate feasibility and select the right method, at the right scale. Build — iterative development with regular checkpoints and measurable objectives. Transfer — documentation, training and handover so the solution lives on without dependency. Discuss your project \u0026rarr; ","externalUrl":null,"permalink":"/en/services/","section":"Services","summary":"","title":"Services","type":"services"},{"content":"","externalUrl":null,"permalink":"/en/tags/","section":"Tags","summary":"","title":"Tags","type":"tags"},{"content":"SubAlpine.IT is an independent consultancy founded by Jérôme Buisine, PhD in Computer Science, offering machine learning, data science, optimisation and software development services.\nThe right method for the right problem: solutions are carefully selected and designed to be stable, relevant and proportionate — without ever exceeding the dimensions of the need.\nMachine Learning \u0026amp; AI Design, training and deployment of models adapted to your data and constraints — from classical learning to deep and reinforcement learning. Data Science Exploration, statistical analysis and modelling of your data to extract reliable insights and support decision-making. Optimisation Metaheuristics and operational research methods to reduce costs, time and resource consumption of your processes. Software Development Design and development of tailor-made applications — back-end, web and mobile — from architecture to deployment, built to last. SubAlpine.IT also builds its own products — StratAlp, Swoop and Meristem. Get in touch \u0026rarr; ","externalUrl":null,"permalink":"/en/","section":"Welcome to SubAlpine.IT!","summary":"","title":"Welcome to SubAlpine.IT!","type":"page"},{"content":" Hi! I'm Jérôme Buisine, PhD in Computer Science and founder of SubAlpine.IT. I'm sensitive to current ecological and climate issues, and my job is to adjust and conceptualise machine learning models in order to meet specific needs — no more, no less. Background and experiences Freelance Data Scientist 2024 - Now. @SubAlpine.IT - France Machine learning \u0026amp; AI solutions Optimisation (metaheuristics, operational research) Data science \u0026amp; statistical analysis Software development (Python backend, PostgreSQL, mobile \u0026amp; web) Scientific consulting \u0026amp; R\u0026amp;D Associate Professor in IT 2022 - 2023 @LISIC - Calais (France) Machine learning methods to optimise time and cost in synthetic imaging Computer graphics research University teaching (computer science, data science) Temporary Lecturer and Research Assistant 2021 - 2022 @LISIC - Calais (France) Research on noise perception in computer-generated images Teaching programming and machine learning PhD in IT 2018 - 2021 @LISIC - Calais (France) Subject: Machine learning methods for taking into consideration noise in computer generated images. Carried out within the ANR PrISE-3D project (Perception, Interaction and 3D Lighting Simulation). MSc in IT (with distinction) 2016 - 2018 @ULCO - Calais (France) Research My research work converges around three axes: image synthesis (Monte-Carlo rendering, noise and artifact removal), deep learning (neural architectures for noise characterisation and detection) and visual perception (eye fixations and perception thresholds). It has led to publications in journals and conferences such as Journal of Vision, Entropy, ICMLA, the ACM Symposium on Applied Perception and the Eurographics Symposium on Rendering, as well as an open-source optimisation library published in the Journal of Open Source Software.\nYou can find my publications on Google Scholar and ResearchGate, my code on GitHub, and more details on my personal website: jeromebuisine.fr.\n","externalUrl":null,"permalink":"/en/story/","section":"Who Am I?","summary":"","title":"Who Am I?","type":"story"}]