HOW TO GET INVOLVED WITH A PROJECT

Sometimes projects are looking for new members. Others will be happy to chat with you about their work. Been invited to join an existing team? Get ready to grow together in our friendly, open community.

SEE PROJECTS LOOKING FOR NEW MEMBERS

BEEN INVITED TO JOIN A PROJECT?

Sign up for a Project Support onboarding to become a member and officially join the team. This is required to use the SPH co-working space and be part of our community.

Join via Onboarding
The content of the project pages is created independently by the SPH Projects.
The work and opinions expressed do not necessarily represent the views of ETH Zurich or ETH Student Project House.

Projects

Filter Projects:

Status

Recruitment

Project-Category

Project-Topics

Filter Makerspace Projects:

Skill-Level

Themes

Ways-to-Create

Start typing to see results and press Enter or hit ESC to close
Antelope

Alumni

Data-driven & Machine Learning

Hardware Engineering & Robotics

Health and wellbeing

Software Engineering

Sports
Join Antelope, where we are redefining human biomechanics, transforming bodies into precision-tuned data engines at a fraction of the cost.Our breakthrough technology, ten times cheaper and more effective than existing solutions, harnesses the power of wearable sensors to predict muscle force — accurately and non-invasively. Discover how our innovative research is not only meeting but surpassing the current challenges, propelling the future of biomechanical assessments into uncharted territories.With Antelope, experience the forefront of biomechanics innovation!...
Find out more >
Autonomous River Cleanup

Alumni

Data-driven & Machine Learning

Environmental Sustainability

Hardware Engineering & Robotics
The Autonomous River Cleanup focuses on developing a river waste monitoring network and an autonomous robotic sorting system to improve river waste recycling.Application of High-tech in Environmental ProtectionThe project aims to transfer and apply high-tech, such as robotics, automation, and artificial intelligence, to the field of environmental protection. As part of our project, we are developing and testing systems to show how such technologies can be used effectively against river waste.Training the Next Generation of EngineersAt the same time, this project focuses on training future engineers. We deliberately present the project as a learning environment where they can realize and test their ideas. Students can apply the knowledge they have learned in lectures in practice and learn how to organize projects and work efficiently in a team.Raising Public AwarenessIn addition, we use our project as a platform to raise public awareness about the global problem of river waste. We regularly organize public events highlighting the challenges of river waste and how the problem can be solved. ...
Find out more >
BrickMachine

Active

Looking for a team member

Data-driven & Machine Learning

Environmental Sustainability

Hardware Engineering & Robotics

Software Engineering
Unsere Vision ist es, den Klemmbausteinmarkt nachhaltiger und effizienter zu gestalten. Weltweit existieren über 1 Billion LEGO-Steine, von denen viele ungenutzt bleiben, da manuelles Sortieren zeit- und kostenintensiv ist. Nach über drei Jahren Entwicklungszeit haben wir einen vollfunktionsfähigen Prototyp einer automatisierten LEGO-Sortiermaschine entwickelt und erfolgreich getestet. Wir bieten nun unseren Service sowohl im B2C-Bereich für private LEGO-Sammlungen als auch im B2B-Umfeld für Händler, Wiederverkäufer und Organisationen an. Das Projekt begann während unseres letzten Schuljahres, als wir LEGO-Steine von Hand sortierten und weiterverkauften, um unser Taschengeld aufzubessern. Dadurch mussten wir schnell feststellen wie nervig und zeitaufwändig die Sortierung von gebrauchten Steinen wirklich ist....
Find out more >
Caterra

Alumni

Agriculture, forestry, food and beverages

Data-driven & Machine Learning

Hardware Engineering & Robotics
We enable herbicide-free precision farming for a sustainable future. Weeding is a significant problem in farming coupled with high costs. Conventionally, herbicides are used to treat the problem, but with more being prohibited every year, as well as a general trend to organic farming, alternatives are needed. While weeds between crop rows can be removed by tractor mounted cultivators, weeds within the rows must be removed by hand. However, with the increasing number of organic farmers, the required workforce is difficult to find and automation is necessary. We propose a small autonomous rover which drives through the field and removes weeds close to the crops using a laser-based module. Any vegetable grown in a row is a possible....
Find out more >
ClimaLinks

Active

Construction, architecture, built environment

Data-driven & Machine Learning
Discover ClimaLinks: Your AI-Powered Weather Relationship Management Solution Tired of being caught off guard by the weather? ClimaLinks is here to change the game. We're a groundbreaking software startup offering a unique Weather Relationship Management (WRM) platform designed to transform the way you plan your tasks: Weather Scoring System: Our AI engine converts forecasts into practical weather scores, customized for your specific needs.Windows of Opportunity Identifier: Easily find the perfect days for any task in the upcoming days.Planner Checker: Evaluate your schedule for weather risks and make the right adjustments to minimize potential disruptions.Schedule Builder: Generate an optimized schedule based on your tasks and ideal weather criteria. Experience the future of weather management and planning with ClimaLinks. Embrace the weather, and never be caught off guard again....
Find out more >
DIGIT SOIL

Alumni

Data-driven & Machine Learning

Environmental Sustainability

Hardware Engineering & Robotics

Software Engineering
Digit Soil is a team of experts in soil science, biotechnology, agronomy, and software engineering that aims to revolutionize agriculture and regenerate soils. We believe that farmers need accurate and directly interpretable information to farm sustainably, starting with an indication of the optimal amount, type, and timing of fertilizer application to improve their efficiency and foster the use of organic fertilizer. At Digit Soil, we have developed a technology that provides farmers with the information they need to optimize the use of both synthetic and organic fertilizers. Our device measures soil enzymatic activity, the biochemical process responsible for releasing nutrients to plants in the soil. By combining this data with current agronomic models and deep-learning approaches, we can provide farmers with the optimal timing, amount, and composition of fertilizer to increase nutrient uptake by crops, leading to a more sustainable and productive farming system....
Find out more >
Ememe

Active

Art, culture and entertainment

Data-driven & Machine Learning
Ememe Town — this is a town where AI characters live freely. With generative AI technology, NPCs (Non-Player Characters) autonomously engage in conversations, take actions, build relationships between characters, and live their everyday lives as they please. You can create a reality drama with AI characters, as if it were the Truman Show, by creating new AI characters living in this town, setting their personalities, and sometimes intervening in their actions. AI automatically generates a daily schedule based on your NPC's personality. They might oversleep, skip meals, rush to school, have silly conversations with friends, come back, stay up a bit late on social media, and then go to sleep... AI generates such daily life of your favorite character. NPC conversations are generated in real-time by LLM. And animations, too! Using our proprietary technology, AI predicts and outputs appropriate facial expressions and body animations from over 1,000 motion data options based on conversations and schedules. Your character will come to life, laugh, dance, and automatically output 3D animations without the need for complex development techniques. You can freely converse with AI characters! By taking on the inner voice of NPCs or becoming other NPCs in conversation, you can influence actions of NPCs. Encourage them to skip school or badmouth their boyfriend... Let’s see what NPCs do! (Don’t go overboard with pranking AI though.) In this town, unexpected events sometimes occur. Meteor collisions, visits from aliens, winning the lottery, murder cases... and you can create these events! Create slightly spicy surprising events and throw NPCs into the whirlwind of confusion. Upload your favorite 3D character models*! Summon your faves to the town as AI NPCs. You can freely set their personalities. Recreate the appearance and essence of your favorite characters! Ememe aims to be a new AI-driven storytelling platform. In the next development phase, we plan to transform Ememe Town an open-world, where a multitude of NPCs coexist. Additionally, we provide users with tools to develop various multiplayer events (game missions) taking place in the town....
Find out more >
Engineers for Science

Alumni

Community empowerment

Data-driven & Machine Learning

Education

Software Engineering
Engineers for Science is an initiative around promoting software engineering best practices in research. We believe that great science requires great software. Unfortunately, many researchers struggle to write effective code, leading to poor software quality, slowing down scientific progress. That's where we come in. We connect researchers with experienced software engineers from the industry who can provide valuable feedback and guidance on best practices for software development. You can learn about some of these principles in our blog. By partnering with Engineers for Science, researchers can ensure that their software is a robust tool for their research and not a hindrance, enabling them to focus on their scientific discoveries. But it's not just about helping researchers. Our initiative also benefits software engineers who are looking to gain experience in mentorship and leadership. By working with researchers, the engineers can gain valuable insights into the latest scientific discoveries and can apply their skills in new and exciting ways. Join Engineers for Science today and help us bridge the gap between science and software engineering. Together, we can create better software and drive scientific progress forward....
Find out more >
exabrain

Active

Data-driven & Machine Learning

Education

AI
exabrain transforms your study materials into hyper-personalized practice. Exercises adjust to your pace and help you master any topic. As you learn, exabrain tracks and adapts to your proficiency, giving you harder questions after correct answers and easier questions after incorrect answers, so you’re always practicing at the right level. exabrain is built for any learning stage, supporting you as a beginner and challenging you as an advanced learner.Built by students, for students....
Find out more >
Floatz

Alumni

Data-driven & Machine Learning

Education

Software Engineering
floatz helps researchers find the most relevant and accurate science through novel AI architectures. 2 million scientific articles are published every year and, to make things worse, the disruptiveness of these articles has dropped by 90% over the last 50 years. How can a researcher find innovation in so much redundant information? The consequence is that researchers spend one year, on average, to write a successful project proposal that unlocks fundings. Practically, a researcher receives hundreds of articles every day from journals, newsletters etc. Most of it is irrelevant and they end up, after many hours, picking what seems useful while looking for novelty or impact. floatz provides a familiar and streamlined user experience. Researchers prompt for what they need and floatz uses AI to deliver relevant science and display it in a way that is easiest to digest. It combines the relevance capabilities of ChatGPT with the precision of scientific search engines like Google Scholar....
Find out more >
Helena.ai

Active

Chemical processes + compounds, biotech and pharmaceuticals

Data-driven & Machine Learning
We aim to improve drug-candidate selection by predicting the effect of drug perturbations on a virtual cell based on single-cell sequencing data. If successful, this should enable the development of more and better drugs.We are currently iterating on several models (autoencoders, drug embeddings, flow matching models) that work together to predict the effect of drug perturbations. Over the next months, we aim to scale up these approaches by several orders of magnitude both in data and compute.We met as students of the computational biology masters @ ETH and bonded during long, shared hours of exam preparation. We both care deeply about improving the grossly inefficient drug discovery process....
Find out more >
Hephaistos

Active

LifeScienceLab

Chemical processes + compounds, biotech and pharmaceuticals

Data-driven & Machine Learning
At Hephaistos, we are revolutionising the chemical, agricultural, and pharmaceutical industries with our innovative approach to enzyme catalysis. In today's rapidly evolving industries, the demand for efficient and sustainable processes is greater than ever. Conventional chemical synthesis, which is at the core of many industrial processes today, often relies on harsh conditions, toxic reagents and requires substantial energy, making it both environmentally and economically inefficient. However, nature itself presents a solution: enzymes. These natural biocatalysts operate under mild conditions, with high specificity and precision, create minimal by-products, while catalysing the same reactions used in conventional chemical synthesis. Despite their immense potential, suitable enzymes for all industrial applications are not always already found in nature. To solve this problem, we have created Hephaistos, a machine-learning (ML) model that learns from atomistic 3D structure data of existing enzymes. By using cutting-edge tools in protein engineering and bioinformatics, our model will be able to generate novel catalytic cores for any desired reaction – even on substrates not found in nature. We do not compete with existing enzymes, but instead create entirely new ones....
Find out more >
hi-robin.ai

Active

Consumer Apps

Data-driven & Machine Learning
robin exists to fix a broken way of finding homes. Today, property search is manual, repetitive, and shallow: filters, endless scrolling, and missed opportunities. robin replaces this with understanding. robin does not search listings. robin understands intent. robin matches humans with properties, not just ads. We are recent ETH graduates. While studying in Zurich, we searched for a suitable home ourselves. Very quickly, we became frustrated by expensive search models, outdated portals, and incomplete, low-quality listings. That experience led to robin....
Find out more >
House Of Charts

Alumni

Data-driven & Machine Learning

Health and wellbeing

Medical devices and diagnostics

Software Engineering
House of Charts revolutionizes hospital workflows by using advanced AI to automate clinical documentation, allowing medical staff to focus on patient care. Their privacy-preserving technology generates discharge reports, medical coding, and other essential documents, seamlessly integrating with existing systems to enhance efficiency across all specialties. ...
Find out more >
Irmos (ex-SeismoHealth)

Alumni

Data-driven & Machine Learning

Hardware Engineering & Robotics

Software Engineering
Maximizing the safe life of bridges and buildings based on data Irmos'software translates vibrations of buildings and bridges due to wind or traffic into structural health metrics for a safer and more sustainable infrastructure. Based on a limited number of low-cost sensors that they developed, the software tracks key damage indicators and supports the decision making for optimial maintenance planning. Their solution attracted key infrastructure owners and top engineeing offices, counting already multiple pilot projects, including an 8-year monitoring contract with the Swiss federal office for the road network (ASTRA). The founders count 4 PhD and 6 MSc degrees in civil engineering, structural health monitoring, data science and robotics and the team is growing from 4 to 8 until the end of 2023....
Find out more >
KAJA AI

Active

Data-driven & Machine Learning

Software Engineering
An autonomous Sales agent that accesses the world's largest dataset of Swiss companies to automatically draft highly personalized outreach emails for your Sales campaigns. We are currently running paid pilots with initial clients to validate our technology and market fit. Our main goal is to scale our data infrastructure to soon cover the entire DACH region and revolutionize B2B sales. We are 3 friends studying Computer Science together at ETH Zurich. We came together to combine our technical skills and build an autonomous agent that automates Sales....
Find out more >
KORA LABS

Alumni

Agriculture, forestry, food and beverages

Data-driven & Machine Learning
KORA LABS provides personalised and sustainable fertilisation plans for any farm.We believe that feeding the world should not destroy limited natural resources like groundwater and agricultural soils. Moreover, 60% of food is produced by farms smaller than 20 hectares. For this reason, we are on a mission to democratise precision agriculture. To do this, we develop agriculture management tools that make sustainable practices a reality and that can be used for farms of any size and type....
Find out more >
Lumia

Alumni

Data-driven & Machine Learning

Social Impact

Software Engineering
Revolutionizing Fair Hiring: Introducing AI-Powered Solution by Lumia! 🚀 Our cutting-edge platform empowers recruiters to make equitable decisions, by becoming aware of and overcoming unconscious biases that affect the hiring process. This is done through Lumia's Fairness Engine which is our one-of-a-kind model that is built on principles of fairness and is able to analyzes past hiring decisions to identify potential biases based on gender, age, race, and more. 📊 Boost diversity and attract top talent with our certification of inclusivity. Join us in shaping a future where every candidate is evaluated fairly. Learn more at our website....
Find out more >
Manukai

Active

Data-driven & Machine Learning

Hardware Engineering & Robotics

Software Engineering
Currently, more than 30% of total manufacturing costs and 50% of a worker’s time are devoted to translating designers’ plans into CNC machine code. At MANUKAI, our mission is to bridge the gap between design and production. To achieve this, we’re creating an intelligent assistant that simplifies the development process within Computer-Aided Manufacturing (CAM) programs. Our system learns from CAM programmers’ preferences, streamlining programming efforts with suggestions. Additionally, our software actively oversees the programming process, continuously optimizing programs to reduce machine manufacturing time and enhance overall quality....
Find out more >
MoonAI

Alumni

Data-driven & Machine Learning

Software Engineering
At MoonAI we believe that the future of VC is decentralised and AI-augmented. We’re building tools to enable this. Our first product is an AI-enabled CRM for VCs to find deals, analyse them and win them. Our product helps you understand your investment hypotheses by finding entrepreneurial talent in the space, identifying key research and searching the internet for related startups. Our next feature will help VCs build decentralised scout networks to find startups before they’re incorporated....
Find out more >
MSMD Robotics

Active

Construction, architecture, built environment

Data-driven & Machine Learning

Hardware Engineering & Robotics
MSMD Robotics are developing a humanoid robot that will be employed in the construction sector to solve the growing issues caused by labour shortages. Working alongside humans, seamlessly integrating into existing workflows, eliminating the need for workers to complete repetitive tasks, at an affordable price. Our proof-of-concept HopeJR arm shows what is currently possible with available resources. We’re set to bringing them to the next level. A grand vision for robotics is what led to the creation of MSMD and is now its drive. Believing that the future of this technology will belong to whoever is bold enough to shape it, we took matters into our own hands. We’ve welcomed new members and are always open to those who are passionate about the topic....
Find out more >
NeuroPrune

Alumni

Data-driven & Machine Learning

Software Engineering
At NeuroPrune, we are inspired by the biological process of neural pruning. In the realm of artificial intelligence, we believe the path to advancement isn't solely through expansion or smarter chatbots but through strategic enhancement and, sometimes, necessary reduction. Our mission is to mature neural networks. This philosophy led us to our initial concept of machine unlearning, introducing anonymity as a service — the idea designed to enable neural networks to forget, thereby protecting user privacy by erasing sensitive data like leaked passwords or personal information. However, our journey didn't stop there. We pivoted towards an underrated yet profoundly impactful area: facilitating conversations between chatbots. Imagine a digital offshoring realm where chatbots, each with unique roles and expertise, communicate with each other to solve problems more efficiently than a single, more advanced AI could. Or instead of having to ask the chatbot multiple times until it returns the answer you want, let another chatbot do that work for you. This approach not only has great use cases, but also makes us think about AI shortcomings, such as bias and hallucinations. Communication and the freedom to structure conversations gives us the framework to counteract those....
Find out more >
Optiverse

Active

Data-driven & Machine Learning

Professional services and consulting

Software Engineering

Web3
Pioneering the Future of Professional Video Interaction Dive into the world of Optiverse, where we harness the power of AI to revolutionize how professionals engage with video content. A haven for executives and organizations, our software is the cornerstone of productivity and insightful learning. At Optiverse, our mission is simple yet ambitious – to transform the overwhelming influx of video information into concise and actionable insights. Our cutting-edge platform reimagines lengthy videos into custom, engaging summaries, saving precious time and boosting productivity. But our innovation doesn't stop at summarization. We empower companies with a comprehensive ecosystem tailored for video knowledge management, equipped with the unique ability to rebuild videos to cater to specific viewer requirements. Proudly standing as the only platform offering video-to-video summarization solely focused on productivity, Optiverse introduces an unparalleled advantage in the market. Our user-friendly interface prioritizes efficiency, ensuring that every second of video content holds significant value without compromising on the essentials. Join us on this ambitious journey to redefine professional video consumption and shape the future of information accessibility. Optiverse – Where every frame counts....
Find out more >
P2P Industries

Alumni

Data-driven & Machine Learning

Hardware Engineering & Robotics

Software Engineering

Web3
Operating swarms of robots in a simple and out-of-the-box way. P2P Industries enables resilient connection, high-bandwidth communication, and abstracted control.We have built networking software that will be open sourced for academia and sold to enterprises.We have secured funding and started refining the networking software to the needs of roboticists. The goal is to be the number one networking extension for robots.The project grew out of a hackathon, more than one year ago. It morphed into an IoT company, and after countless hours in SPH of building our software and an automatic plant care demo, we have started to use the stack for robots.The team has known each other for over 3 years in CS studies....
Find out more >
Qcomply

Active

Data-driven & Machine Learning

Professional services and consulting

Software Engineering
So far, we’ve built an early MVP that automatically scans systems to find where cryptography is used and flags what isn’t ready for a post-quantum world.Our main goal is to make cryptographic discovery and quantum-safe migration simple enough that any security team can handle it without deep crypto expertise.We’re now working with several organisations to run early pilots and refine the product. In the long run, we want to become the easiest way for companies to stay quantum-safe. The project started when we realised that most companies have no clear overview of their cryptographic assets, even though upcoming regulations will soon require it.We teamed up to build a tool that removes the manual pain and makes quantum readiness actually achievable for real organisations....
Find out more >
Rapidata

Active

Data-driven & Machine Learning
Rapidata solves a fundamental problem faced by the modern AI industry. AI tools and systems, like ChatGPT, autonomous driving or magic erase, rely on massive amounts of data and the secret of the industry is that this data needs to be filtered, annotated and validated by enormous amounts of repetitive human labor in low cost countries. This is slow, expensive, and most of all, not scalable. Rapidata solves this by not relying on a fixed workforce, instead we have an existing network of Millions of People around the world that we can access through the same technology as digital ads. This allows us to offer 1000x faster human-verified data processing at scale, while being more affordable than the competition. At the same time we offer a revolutionary new way to monetize free web services!...
Find out more >
RiverLabs

Active

Data-driven & Machine Learning

Hardware Engineering & Robotics
Two billion people currently source their drinking water from rivers, yet freshwater ecosystems are the most degraded in the world. Basic information about our freshwater systems, such as water temperature, quantity, and clarity, are missing because these ecosystems are dynamic and because of the lack of affordable measurement systems. At RiverLabs, we collect and provide this freshwater data using a network of self-made sensors, that cost less than traditional sensors on the market. If you are interested in partnering with RiverLabs and if you would like to set up a sensor network in your watershed, then please get in touch. ...
Find out more >
spAIce

Alumni

Aerospace, automotive and nautical industries

Data-driven & Machine Learning

Hardware Engineering & Robotics

Software Engineering
Enabling in-orbit services at scale with native AI in space With the increasing frequency of satellite launches and impending space debris regulations, SPAICE is meeting the need for effective in-space servicing. Our solutions can enhance satellite lifespans by 50%, boost their revenue generation by 50%, and reduce collision risks by 20%. Traditional methods, limited by manual imprecise terrestrial controls, are inaccessible to 90% of the satellite market due to their prohibitive costs. SPAICE pioneered leveraging Generative AI and Deep Reinforcement Learning to develop the first fully autonomous solution for In-Orbit Services (IOS). Our proprietary software tools (under patent filing) enable seamless navigation and multitasking robotic services and integrate with existing satellite systems to improve their capabilities. Our software is delivered through a plug-and-play servicing platform including sensors, robot arms, and a computing unit. As a modular payload, our platform is ready to be integrated into any existing satellite, transforming it into a functional in-orbit servicer. Our positioning is centered around being a technological enabler, providing the missing piece to allow IOS to be applied at scale. Our team has developed the world's largest photorealistic dataset of orbiting assets, comprising over 1 million images. By learning to optimize universal reward functions, SPAICE's technology excels in understanding perception features and control policies from data patterns, directly through interaction with the environment. This innovation is scalable across multiple use cases and adaptable to varying targets and mission requirements, overcoming the main limitation of current IOS prototypes, whose design is tied to a particular target satellite and mission, thus being unscalable. Founded by a top-tier team from prestigious institutions including Imperial College London, ETH Zurich, and Oxford, SPAICE is at the forefront of space technology. We're currently validating our solutions through Proof of Concepts (PoCs) with leading European space entities and have an exciting lineup of projects in our pipeline....
Find out more >
Synature

Active

Agriculture, forestry, food and beverages

Data-driven & Machine Learning

Social Impact
We are two co-founders developing bio-acoustic technologies to measure the state of biodiversity. The aim of Synature is to make biodiversity quantifiable and to therefore enable a better co-existence of humans and nature. This involves recording sound and using AI and signal processing to analyse the gathered data. Synature has a holistic offering including smart microphones that send audio recordings from the field, automated analysis of the gathered data and a dashboard that displays the data in an easy-to-understand way.Highlights of Synature include a paid pilot project with the Swiss National Park and the Yellowstone National Park focusing on acoustic wolf detection, media appearances on national television, market exploration in South Africa and India, paid pilot projects in Valais and Paris and multiple awards....
Find out more >