Practice Science Labs

160+ virtual science experiments with real measurements, any school, any device

Hands-on experimentation is essential to learning science, and most students get very little of it: physical labs need capital, space, safety controls, and timetable slots, and even where they exist, large classes and single-run experiments leave no room to repeat an attempt or learn from a failure. Constructor Practice – Science Labs is a cloud-based platform of 160+ interactive virtual experimen

bookmark_border

Bookmark

play_arrow

Overview

Information on this page is provided by the innovator and has not been evaluated by HundrED.

Updated August 2026

Web presence

2018

Established

7

Countries

All students

Target group

That access to a laboratory stops determining whether a student gets to do science. Practical work is currently rationed by capital, space, safety, and timetable – which means it is rationed by how wealthy a school is and how far from a city it sits. We would like every student, in a rural school or a well-funded one, to be able to run a real experiment, take their own measurements, get the error analysis wrong, and run it again. We would also like practical competence to be assessed as practical competence. Most systems test whether a student remembers a concept, because assessing the experimental process at scale has been impractical. If how a student sets up a measurement and reasons about its error can be evaluated as reliably as a multiple-choice answer, then teaching can afford to spend time on the part that matters – and 'doing science' can mean doing it, rather than watching it done.

About the innovation

Why did you create this innovation?

Hands-on experimentation is essential to learning science, and most students never get it.

Physical labs demand capital, dedicated space, safety controls, and scheduling capacity. In under-resourced and remote settings, students therefore rarely run a genuine experiment at all. Even where labs exist, large classes, limited equipment, and single-run sessions cut individual practice to almost nothing: a student who mishandles a measurement does not get a second attempt, and a failed attempt is where most of the learning was.

The available substitutes do not close the gap. Videos and static simulations ask students to observe rather than experiment. They supply the expected result rather than a measurement, so there is no authentic data to work with and no error to interpret – and error analysis is a large part of what doing science actually is.

The result is a persistent gap between understanding a concept and being able to practice it, distributed unequally: the students with the least access to a lab are the ones for whom the theory stays most abstract. Teachers, meanwhile, carry a heavy manual marking load for the practical work that does happen.

We built Science Labs so that a student with a browser and an existing device can run a real experiment – take unique measurements, compute with real formulas, get the error analysis wrong and do it again – as many times as it takes, and so that the process of doing so can be assessed rather than only the recall.

What does your innovation look like in practice?

A teacher sets an experiment as an assignment, in a browser, on the devices the school already has. There is nothing to install and no equipment to book.

The student opens the experiment and works through it as they would in a physical lab: set up the apparatus, take readings from realistic instrumentation, record what they measured. The measurements are unique to that run, so there is no answer to copy and no expected result to reverse-engineer. The student then computes with real scientific formulas and carries out the error analysis – theoretical, observational, and instrumental – which is where most of the scientific reasoning happens.

If it goes wrong, they run it again. This is what a physical lab cannot usually offer: repetition is free, so a failed attempt becomes something to diagnose rather than a lost lesson.

The teacher sees submissions with AI-assisted grading and progress tracking, and the assessment covers the experimental process – how the measurement was taken and reasoned about – not only a conceptual quiz at the end. Because the platform carries 160+ experiments across mechanics, chemistry, biology, molecular biology, thermodynamics, electromagnetism, optics, physiology (EEG, ECG, EMG), and junior science, aligned to IB, AP, NGSS, GCSE, and university programs, a department can map it onto the syllabus it already teaches rather than teaching around the tool. Access runs through LMS single sign-on, so students use their existing school login.

How has it been spreading?

Science Labs grew out of the VRLab Academy platform, founded in 2018, which reached 163,097 registered users, of whom 141,589 were learners, across a footprint that now spans 40+ countries. Constructor Tech acquired it in 2025 and rebuilt it as Constructor Practice – Science Labs; migration from the legacy platform is ongoing, and the new platform currently has 12 institutional customers and 2,531 active users, with annualised project revenue of about US$360K.

The catalogue has been the main engine of spread: 160+ experiments live across mechanics, chemistry, biology, molecular biology, thermodynamics, electromagnetism, optics, physiology, and junior science, with microbiology and first aid in development, and alignment to IB, AP, NGSS, GCSE, and university programs. A department can adopt it against the syllabus it already teaches, which is why adoption tends to start with one course and widen.

The platform has been independently reviewed by LearnPlatform by Instructure as meeting ESSA Level IV ("Demonstrates a Rationale"), which matters for United States district procurement. It is distributed as part of Constructor Tech (2,000+ partners and customers, 2.5M+ platform users).

How have you modified or added to your innovation?

The platform has changed substantially since the 2018 VRLab Academy release. The experiment catalogue has grown to 160+ across mechanics, chemistry, biology, molecular biology, thermodynamics, electromagnetism, optics, physiology (EEG, ECG, EMG), and junior science, with microbiology and first aid currently in development, and alignment added to IB, AP, NGSS, GCSE, and university programs.

Following the 2025 acquisition by Constructor Tech it has been rebuilt as Constructor Practice – Science Labs, with migration from the legacy platform under way. The rebuild added assignments, AI-assisted grading, and progress tracking so that the experimental process itself is assessed rather than only conceptual recall, and LMS single sign-on so students reach experiments through their existing school login rather than a separate account.

The platform has also been submitted for independent review, and was assessed by LearnPlatform by Instructure as meeting ESSA Level IV ("Demonstrates a Rationale").

If I want to try it, what should I do?

Start with the product page: constructor.tech/products/learning/practice/science-labs, and watch the product video – it shows an experiment being run, which explains the difference from a simulation faster than any description.

If you are a teacher: pick one experiment from your current syllabus – something your class is covering in the next two weeks – and run it yourself first, end to end, including the error analysis. That tells you in twenty minutes whether the measurement work matches what you expect of your students. Then set it as an assignment for one class and look at what the submissions show you about their reasoning, not just their answers.

If you are a department or an institution: ask for a pilot on one course. Check the catalogue against your syllabus (IB, AP, NGSS, GCSE or your university's own program), confirm LMS single sign-on with your IT team so students use existing logins, and run the pilot for a term with the AI-assisted grading turned on so you can judge the marking-load reduction as well as the learning.

To talk to us: get in touch through constructor.tech and ask about a Science Labs pilot.

Implementation steps

Run one experiment yourself, end to end

Choose an experiment your class is covering in the next two weeks and complete it yourself, including the measurement and the error analysis. Twenty minutes tells you whether the measurement work matches what you expect of your students – and it is the fastest way to see the difference from a video or a conceptual simulation, because your readings are generated for your run.

Map the catalogue against your syllabus

Check the 160+ experiments against what you teach: alignments exist for IB, AP, NGSS, GCSE, and university programs across mechanics, chemistry, biology, molecular biology, thermodynamics, electromagnetism, optics, physiology (EEG, ECG, EMG), and junior science. Identify the courses with the weakest current practical provision – those are where the platform earns its place fastest.

Connect it to your LMS

Have IT set up single sign-on so students reach experiments through their existing school login. This is worth doing before the first assignment rather than after: separate accounts are the main source of lost time in the first week, and the platform needs no installation on student devices.

Assign one experiment to one class

Set a single experiment as an assignment for one class rather than rolling out across a department. Because every student's measurements are unique, there is nothing to copy – expect submissions to differ, and look at how students reasoned about their error rather than only at whether the final value is right.

Use the grading and progress data to judge a pilot

Turn on AI-assisted grading and track a term. Judge two things separately: whether students' practical reasoning improves, and how much marking time the department recovers – the second is what usually funds the decision to widen. Then compare against the practical provision the course had before.