You bring the question.
Drag and drop your data file. Set your goal and variables. You contribute the context and knowledge of your field; no coding or statistics expertise is required.
Scientific intelligence for data-driven decisions
From data to
defensible
decisions.
OptiMod automates scientific data analysis, turning complex data into clear, defensible evidence. Understand the findings, the uncertainty and the reasoning—for decisions in research, business and public policy.
No coding. No statistics expertise required.
OptiMod / Findings
Capital Bikeshare731 daily observations
50% lower
daily demand
Adjusted association for light rain or snow compared with clear weather.
95% confidence interval · 38%–60% lower
Research. Support. Shared ambition.Explore our network
Scientific scrutiny, within reach
Results should earn trust
before they inform decisions.
Scientific data analysis should be within everyone’s reach.
Organisations collect more data than their statistical specialists can examine. OptiMod puts that expertise into software, so researchers and domain teams can investigate their own questions.
General-purpose AI language models are not a substitute for a scientific validation system. OptiMod evaluates the data and model assumptions, compares suitable methods and documents the calculations behind each finding.
Our ambition is a standard of evidence that lets others examine how a conclusion was reached. Explore the applications
OptiMod takes it from there.
Data preparation, method requirements, assumption checks, model comparison and analysis—all handled automatically, with the reasoning recorded.
You receive the evidence.
Understand the findings and their uncertainty. Keep the Scientific Report with the methods, named tests and analytical decisions behind the result.
02 / The OptiMod approach
Scientific knowledge,
built into the workflow.
OptiMod combines mathematical and logical reasoning with targeted AI assistance. It evaluates your actual data, tests the requirements of candidate methods and selects the best-supported model for your question. Automatically.
01
Understand
the data.
OptiMod examines data quality, variable types, distributions and relationships. It prepares the dataset and records each change.
02
Establish
what fits.
Mathematical and logical checks establish which methods fit the data and the question. Model assumptions are part of this evaluation from the start.
03
Select with
evidence.
OptiMod compares eligible models using fit, diagnostic evidence and the analytical objective. The selected model comes with a documented scientific rationale.
04
Explain.
Document. Deliver.
Estimates, uncertainty and interpretation stay connected to the calculations behind them. Each finding can be traced back to its data, methods and checks.
From a dataset and a question to a documented scientific result.Inside the scientific core
Research depth / The replication challenge
Make rigorous analysis
part of the routine.
70%+
of researchers surveyed had tried and failed to reproduce another scientist’s experiments.
Survey of 1,576 researchers · Nature, 2016Our ambition is to make scientific scrutiny routine. When statistical expertise is available within the workflow, researchers can move from experiment to finding with the evidence that reviewers need to examine it.
Rigour as
the default.
Researchers gain access to the statistical checks their work calls for, even when their department has limited specialist capacity.
Evidence others
can scrutinise.
Collaborators and reviewers can examine assumptions, uncertainty and the rationale for a finding, then judge which conclusions the evidence supports.
Results others
can build on.
The Scientific Report connects each finding to its analytical choices. Other teams can revisit the work and use it as a basis for further research.
Rules and calculations.
Mathematical and logical checks establish which methods fit the data. Model choice is backed by diagnostic and statistical evidence.
Targeted AI assistance.
AI helps make the process efficient and the findings understandable. Explanations stay connected to the calculations and checks behind them.
The product experience
Complex science.
A remarkably easy experience.
Drag and drop your data file and set your goal. OptiMod handles the scientific analysis automatically, then explains the findings and documents how it reached them.
Daily demand analysisExample datasetDATASET
Add your data.
Drag in your dataset. No coding or statistics expertise needed.
Drop your dataset here
CSV, Excel & more · your analysis starts here
✓ 731 records recognized · source retainedTry it: drag the example file here, or click to use it.A first look at your data
| Date | Daily rides | Weather |
|---|---|---|
| 01 Jan 2011 | 985 | Mist / cloudy |
| 02 Jan 2011 | 801 | Mist / cloudy |
| 03 Jan 2011 | 1,349 | Clear |
Three rows from the original public dataset. Your source file stays unchanged.
ANALYSIS SETUP
Choose your analysis goal.
Choose what to explain and which variables to consider. OptiMod determines the statistical approach.
EXPLAIN A RELATIONSHIP
How is weather associated
with bike-sharing demand?
What to explainDaily ridesNumber of rentals per day
What to considerWeather & calendar patternsWeather, temperature, humidity, wind and time
AUTOMATIC ANALYSIS
Analysis complete.
OptiMod evaluates the data and method requirements, compares eligible models and computes the results.
Data checks & preparation
731 records checked. Leakage and redundant columns excluded.
Method checks & model selection
Count models compared against the data and diagnostic evidence.
Results & scientific documentation
Confidence intervals, sensitivity checks and remaining limits recorded.
Scientific checks recorded
MAIN FINDING
Lower demand on rainy days.
Weather is associated with daily demand, after accounting for other variables.
LIGHT RAIN / SNOW VS CLEAR WEATHER
50%lower daily demand
Estimated association in this dataset.
95% interval: 38%–60% lower.
Expected demand, relative to clear weather
Model-based comparison; other variables held fixed. Clear weather = 100.
i
This is an association, not proof of cause and effect. Model limits remain part of the analysis.
SCIENTIFIC REPORT
Your analysis. Fully documented.
From model assumptions to named statistical tests: the scientific record behind your result.
SCIENTIFIC REPORTDaily bike-sharing
demand.
Data. Methods. Evidence.
Preparing the report…
Example report ready
- 01What was analysed
The question, source data and preparation.
- 02Assumptions & exact statistical tests
Test names, statistics, p-values and interpretation.
- 03What the evidence supports
Findings, uncertainty and limitations.
Keep and share the record independently of access to OptiMod.
Example dataset · ready to exploreDataset
Drop in your data. OptiMod handles the scientific analysis once you set the goal.
Real public dataOpen the full product tour
Interface preview, in final development. The example uses real public data and independently computed results. Playback illustrates the analysis and report export workflow.
731 days of recorded bike-sharing demand in Washington, DC, from January 2011 to December 2012.
- Original source
- UCI Bike Sharing · day.csv · 16 columns
- Missing values / duplicate days
- 0 / 0
- Source integrity
- Original file retained; SHA-256 recorded in the report
The preview uses readable column labels. The downloadable CSV preserves the original names and normalized units.
Download the original CSVAn explanatory analysis of conditional relationships, using observed daily weather and calendar data.
- Outcome
cnt· daily rental count- Source variables
- Weather category, temperature, humidity, wind, holiday, weekday and date.
- Calendar adjustment
- Month, weekday and a continuous time trend; a quadratic temperature term.
- Excluded from predictors
casualandregisteredalready sum to the outcome;instantis an identifier;atempduplicates temperature information.
The question and setup are fixed for this example. This is not a future-weather forecast or an estimate of causal effects.
The checks make scientific choices visible. They also retain findings that need judgement.
Data preparation
All 731 rows are retained. No missing values are imputed and no outliers are silently removed. Two target-leakage columns and a redundant temperature variable are excluded.
One humidity reading is zero (10 March 2011). It is flagged and the analysis is repeated without it as a sensitivity check. Temperature stays in normalized source units because archive metadata conflict.
Model comparison
Poisson and Negative Binomial (NB2) are fitted to 547 training days. Training AIC is 66,197.6 for Poisson and 8,815.7 for NB2. Poisson dispersion is 109.20, indicating marked extra variation relative to its variance assumption.
NB2 is selected for the association analysis. Poisson has the lower prediction error in the 184-day holdout: 689 versus 863 rides/day. Model selection follows the question, not a universal winner.
Remaining review points
Coefficient intervals use 7-lag HAC covariance. Residual lag-1 correlation remains 0.387; this adjustment does not remove model limitations. Weather category 3 contains only 21 days. Study design, measurement and transfer to other settings still require review.
Light rain or snow is associated with a 50.4% lower expected daily count than clear weather, conditional on the included variables.
- Estimated count ratio
- 0.496 · clear weather = 1
- 95% confidence interval
- 0.399 to 0.617 · 38.3% to 60.1% lower demand
- Analysis sample
- Full dataset · 731 days · Negative Binomial regression
- Uncertainty
- Approximate pointwise HAC interval · 7 daily lags
The chart shows adjusted count ratios, not raw average demand. The mist/cloudy estimate is 0.910. These exploratory associations have no multiplicity adjustment and do not establish causation.
Evaluation is a separate question
Held-out prediction error uses a model fitted to the first 547 days. Its NB2 MAE is 863 rides/day and RMSE is 1,123; Poisson has a lower MAE of 689. The 184 held-out days use realized weather, not weather forecasts.
An exact record of the calculation: the question checked, the procedure used, its numerical result and the interpretation.
| Question / assumption | Test or diagnostic | Result |
|---|---|---|
| Residual autocorrelation through 7 daily lags | Ljung–Box Q df = 7 | Q = 251.9104 p = 1.0863 × 10−50 Dependence detected |
| Residual autocorrelation through 14 daily lags | Ljung–Box Q df = 14 | Q = 258.9912 p = 3.9574 × 10−47 Dependence detected |
| Can the mean-model coefficients be estimated? | Design-matrix rank | 26 of 26 columns Full rank · no p-value |
| Is Poisson variance consistent with the observed variation? | Pearson dispersion Training model | 109.20 Marked overdispersion · diagnostic, not a formal p-value test |
The interpretation is part of the record.
The Ljung–Box null is no residual autocorrelation through the stated lag. Both reference checks indicate remaining dependence. Here they are descriptive residual checks using model_df = 0, not definitive NB2 specification tests.
Seven-lag HAC covariance is used for coefficient uncertainty. It does not remove the dependence or guarantee that the model is correctly specified. The report preserves this finding alongside the model choice and results.
Export Analysis Report
The reasoning travels
with the result.
The Scientific Report records the model choice, the assumptions tested and the uncertainty behind the finding. Keep the PDF and share it independently of access to OptiMod.
- Methods
- Data preparation, variable definitions and the rationale for the selected model.
- Checks
- Named statistical tests, test statistics, p-values and diagnostic findings.
- Interpretation
- Estimates, uncertainty and limits that matter when the result informs a decision.
Download the example reproducibility package
Includes source data, analysis code and dependency versions.
One scientific core,
two ways to use it.
Individuals gain access to scientific analysis without needing statistical expertise. For platform providers, the opportunity is to bring the same capability into software, data workflows and AI systems.
For individuals and teams.
Upload your data and set your goal. OptiMod handles the statistical work automatically, evaluating the data and method requirements to select a suitable analysis. Get understandable findings and a Scientific Report.
Your analysis
dataset.csvYour data
Your goal
Understand what influences my outcome.
OptiMod handles the analysis
Data preparation, assumption checks and evidence-led model selection.
Findings you can understand.
What the results mean, how certain they are and what to consider.
Scientific ReportThe reasoning behind your results
A capability for systems.
Connect OptiMod’s scientific core to analytics platforms, enterprise software, research systems and AI assistants. Explore bringing automated analysis and traceable findings into your own products and workflows.
AI assistant
Integration concept
You
dataset.csv
Analyse these customer data. Which factors are associated with retention?
In the backgroundOptiMod runs the scientific analysis
Assistant
Longer onboarding times are associated with lower retention after accounting for the recorded customer characteristics. I’ve attached the estimates, uncertainty and model checks.
Scientific ReportPDF attachment · Methods, checks and results
Potential applications
Different questions.
A common need for evidence.
Research, organisations and public institutions face different decisions. Each needs an analysis whose method, uncertainty and limitations can be understood.
Does an observed difference remain after accounting for relevant variables?
Turn experimental data into scientifically grounded findings. OptiMod handles model evaluation, statistical checks and interpretation, with a clear analytical record for collaborators and publication.
- Decision
- Which findings warrant further study?
- Evidence to review
- Adjusted estimates, model assumptions and uncertainty.
Which molecular or patient characteristics are associated with an outcome?
Structure exploratory analyses and prioritise hypotheses for further investigation. Pharma Intelligence provides an active research setting for this direction.
- Decision
- Which relationships warrant experimental follow-up?
- Evidence to review
- Candidate explanations, diagnostics and open questions.
Which customer and market factors are associated with revenue or retention?
Compare explanations, examine uncertainty and bring documented evidence into commercial planning.
- Decision
- Which retention initiatives deserve a closer look?
- Evidence to review
- Competing explanations and the uncertainty of each estimate.
Which process conditions are associated with variation in quality?
Identify relationships worth investigating and give technical teams a shared statistical basis for process reviews.
- Decision
- Which conditions should engineers investigate first?
- Evidence to review
- Process associations, variability and diagnostic findings.
Where should a nation invest in flood protection?
Analyse flood exposure, population vulnerability and infrastructure data to inform national investment priorities. Make the assumptions, uncertainty and reasoning behind each conclusion explicit—so decisions affecting lives, critical infrastructure and public resources can be scrutinised and explained.
- Decision
- Which regions warrant priority for flood protection?
- Evidence to review
- Exposure, vulnerability, model assumptions and uncertainty.
Which characteristics are associated with changes in measured risk?
Document estimates, model choices and uncertainty so analytical conclusions can be examined in internal review.
- Decision
- Which changes in a risk estimate need closer review?
- Evidence to review
- Model choices, estimates and the limits of the available data.
How can a platform attach scientific evidence to its answers about data?
Explore embedding method evaluation, assumption checks and documented statistical findings into the software people already use. Give users a clear explanation and the scientific record behind it.
- Opportunity
- Bring scientific analysis into an existing data product.
- Evidence to review
- The method used, checks performed and limits of the finding.
Research & development
Built in dialogue
with science.
2025 — 2028Pharma
COLLABORATIVE RESEARCH & DEVELOPMENT
Pharma
Intelligence
COLLABORATIVE RESEARCH & DEVELOPMENTACTIVE RESEARCH PROJECTFrom statistical methods
From statistical methods
to pharmaceutical research.
OptiMod emerged from Leibniz University Hannover’s ITE. Today, it leads Pharma Intelligence, connecting scientific software development with AI-assisted pharmaceutical research.
Consortium: OptiMod · Leibniz University Hannover · Soranos IT-Service
Inside the project Funded by the European Regional Development Fund (ERDF) and the State of Lower Saxony.
The founders
Scientific depth.
Entrepreneurial drive.
Founded by brothers Ylli and Drini Morina, bringing business development and quantitative research into one team.
Ylli Morina
CO-FOUNDER & CEOStudied Computer Science at Georg-August University of Göttingen. He leads OptiMod’s business development and strategic partnerships, turning its scientific foundation into a clear product and business opportunity.
Connect with the team
Dr. rer. nat. Drini Morina
CO-FOUNDER & CTOStudied Statistics at Georg-August University of Göttingen and is a researcher at Leibniz University Hannover. His work connects data science, innovation research and statistical methodology.
Academic backgroundFor institutions
Your data.
Your control.
For institutions with demanding privacy, security and data sovereignty requirements. OptiMod supports hosting in Germany and deployment entirely within your organisation’s own infrastructure.
Discuss institutional requirementsHosting in Germany.
Hosted in Germany, with GDPR-compliant data processing. OptiMod supports institutions with high requirements for privacy, security and control over their data.
No training on your data.
Your data is never used to train AI models. When your analysis is complete, the data used to produce the results is deleted from the servers.
On your own infrastructure.
OptiMod can run on your own servers or computers. For sensitive environments, analysis stays within your infrastructure and your data stays with you.