Questions people ask
How Elnora works, what it does in a lab, and where your data lives.
What is Elnora?
Elnora is the AI agent for lab work. It plans experiments, writes and optimises protocols, searches the scientific literature and databases, analyses results and draws figures, and remembers how your lab works. Give it a goal and it carries the work through to a file your team can use.
What work can an AI lab agent take on?
Elnora writes or optimises a protocol, reviews the literature, plans the next experiment with the right controls and sample size, analyses a run, fits a dose response from plate readings, reads an instrument export, works out why something failed, and writes the results up as a document, a deck or a workbook.
How does an AI lab agent fit into my team?
You reach Elnora wherever your team already works: a web application, the command line, a hosted MCP server so other agents such as Claude Code can call it, and a REST API. It reads 49 file extensions and writes 12 formats, from spreadsheets and Word documents to instrument exports, so most teams start by uploading the files they already have.
How does an AI lab agent keep research data secure?
Elnora holds an ISO/IEC 27001:2022 certification and a SOC 2 Type 2 attestation. Your data is kept apart from other organisations', role based access controls who sees what, and the protocols and results belong to you.
How do I get started?
Book a demo at elnora.ai/demo. We go through the work you want an agent for, show you Elnora live, and send you a price for your team. Onboarding is included, and teams are usually working with it within days of saying go.
Show all 25 questionsShow fewer questions
What separates an AI lab agent from Claude or Codex?
Elnora arrives set up. From the moment you log in it has the scientific tools, the databases, the knowledge base and the certifications already in place, reachable through the web app, the command line, MCP and the API, so your team spends its time on the science and sees a result on the first task.
How does an AI research agent differ from an assistant?
Elnora works from what your lab already knows. Give it your protocols, results and notes, including the runs that failed, and it reads them before it writes anything new and keeps the standing decisions in memory. It carries a job through, searching the literature, running the statistics in an isolated sandbox, drawing the figure and saving the file.
Which sciences can an AI lab agent work in?
Elnora works in any lab. Biology and drug discovery are the largest use, and electrochemistry and materials science are live today, including impedance analysis from raw potentiostat files, microstructure measurements from tomography, and solid oxide cell thermodynamics. Chemistry, analytical and engineering labs use the same planning, analysis and knowledge base.
Can an AI agent analyse experimental data?
Yes. Elnora profiles the data first, chooses the test that suits its shape, checks the assumptions, and runs it in an isolated sandbox. That covers group comparisons, dose response curves with confidence intervals, survival analysis, mixed and Bayesian models, and figures saved to your workspace under the name you choose.
Does an AI lab agent learn from failed experiments?
Yes, when you give Elnora the records. Upload the failed runs and it reads them before planning the next one, and what it writes up goes back into your knowledge base. Ask it afterwards what to use for the next run and it answers with the parameters those failures produced.
Who does what, the AI agent or my scientists?
Elnora plans the experiment, writes the protocol and analyses what comes back. Your scientists review every output and do the bench work, so a qualified person signs off before anything is used.
Which teams is an AI lab agent built for?
Elnora is built for teams with a lab. Pharmaceutical companies, biotech from seed stage onwards, contract research organisations, and chemistry, materials and electrochemistry groups. If your team spends its week planning experiments, writing protocols, reading papers and analysing results, that is the work it takes on.
How long does it take to get an agent running?
You can start the moment the contracts are signed. Building out your knowledge base takes a day or two, and Elnora itself is ready now, so the only thing we are waiting on is you.
Do we need an IT team to run an agent?
We set Elnora up and run it, so your team logs in and works.
Do I need to reformat or prepare my data first?
Send the files as they are, in whatever shape your lab keeps them, and Elnora works with them.
Will my team actually use it?
Elnora is simple to use, and teams pick it up in days or weeks depending on the size of the organisation, because people see a result on their own work straight away.
Does an AI agent make us an AI native biotech?
Yes, because it changes the work itself. We put Elnora into the systems your team already uses, so an agent does real work at each step of research, from planning an experiment to writing up what came back.
Will AI take my job?
Elnora makes you a better scientist. The work you always wanted to do but never had the time or the technique for is open to you now, and it takes on the research and writing that fill a week, so a team runs more experiments and spends its judgement where judgement is needed.
How does an AI vendor prove its compliance standards?
Elnora’s Trust Center carries the certificates, the policies and the current control status, the security overview sets out how your data is handled, and the SOC 2 Type 2 report goes out on request.
Is lab data sold or used to train AI models?
Your data is yours, and the contract with Elnora says so. We build agents on models other people make, your data is never used to train any of them, and every provider is under an agreement that prohibits it.
What AI models does a lab agent run on?
Elnora runs on frontier models from the leading AI labs and on open-weight models, choosing the one that suits each task and failing over between them, so a change of provider never changes what your team can ask for.
What happens to our knowledge base if we leave?
You take everything with you. Elnora exports your protocols, your results and your knowledge base in a form you can use somewhere else, and we are still rooting for the science.
Why work with a startup vendor during a long procurement?
Elnora moves at your speed. You get the people who build the product, changes in days, and a company whose priority is that your work succeeds.
My project is niche. Can you build what it needs?
Yes. Our engineers work with your team directly to get Elnora doing the specific thing your project needs, and Elnora is in active development, with each batch of new capability published in the changelog.
How much does an AI agent for the lab cost?
An annual platform fee, and credits on top for the work Elnora does. The fee is set by the size of your organisation, and it covers every scientist in the lab. The full breakdown is at elnora.ai/pricing.
Ask the rest on a demo, or send us a message.