Workshops
“Everyone on the team tries AI differently, and nobody knows exactly what we are allowed to do.”
Know where to start with AI. And what you are allowed to do.
Open workshops and in-house dates for chemistry, pharma and biotech. On real spectra and data, with an honest answer on what works today.
30 minutes on Microsoft Teams with the CovaSyn team. If it is not a fit, we will say so.
- November 2026
- AI readinesshalf day
- December 2026
- AI for chemiststwo days
- Data protection and AItwo days
- January 2027
- Structure elucidation with AItwo days
- AI readinesshalf day
- February 2027
- Stability per ICH Q1two days
- ICH M7 in silicotwo days
- March 2027
- GxP and self-hostedtwo days
If nothing changes
What it costs when everyone tries AI differently.
- Everyone tries something different, and the results cannot be compared.
- Uncertainty about GDPR, the AI Act and GxP holds back good ideas.
- Pilots start without a clear goal and end without a result.
How we help
How our workshops work.
Method part and tool part kept separate, fixed dates, open prices, bookable online.
- Save time
- Your team learns in one session what would otherwise take months of trial and error.
- Save costs
- You invest in the projects that pay off instead of pilots without results.
Proof
Measured, not claimed.
- AI models alone
- 14 to 41%
- With CovaSyn tools
- 76 to 92%
New
AI readiness workshop
Where do we start, what is possible, what are we allowed to do? In half a day, you settle the questions that stall every AI rollout in chemistry, pharma and biotech at the outset.
- Where to start: which tasks in your lab are worth AI today and which are not
- What is possible: what AI does reliably in chemistry today and where it gets things wrong
- What am I allowed to do: GDPR, the EU AI Act and GxP as the framework for your project
- Next step: a concrete plan for what your team starts with
Topic module
AI in the GxP lab
What AI can do in the regulated lab today, what the EU GMP Annex 22 draft requires and how to classify your own use cases. Orientation for QA, QC, analytical and validation teams, not advice.
- For whom
- QA, QC, analytical, validation and lab leadership in pharma, CDMOs and chemistry
- Formats
- Online compact and in-house
- Led by
- CovaSyn team from chemistry, analytics and software
- 01What AI in the lab can do today and where its limits are
- 02The regulatory framework at a glance: Annex 11, the Annex 22 draft, the EU AI Act, GAMP 5 (orientation, not advice)
- 03Static models, LLMs and the principle "the LLM plans, the tools compute"
- 04Live demo: a traceable evaluation workflow with CovaSyn tools and a human in the loop
- 05Exercise: inventory your own use cases and classify them by criticality (Readiness Check as worksheet)
- 06Roadmap: what the lab can put in place now
The modules are shortened or extended depending on the format.
Background reading: AI in the GxP Lab and EU GMP Annex 22 and the Annex 22 Readiness Check.
Annex 22 is available as a draft (as of 5 October 2026). This content is not legal, regulatory or compliance advice.
Hands-on, from first steps to regulated operation
The dates: the right workshop for every step
From the first overview to operation in a regulated environment. Practical, on real data, in Leipzig or online.
19 Nov 2026
Half day online (9am to 1pm)online
AI readiness workshop: what works, what does not, what it costs
The entry point, no prior knowledge needed. On real spectra and chromatograms instead of text snippets, with an honest answer on where AI still gets things wrong.
Method
- How language models work and why they fail in chemistry
- How to tell a sound answer from an invented one
- Which lab tasks are worth it today and which are not
Hands-on with tools
- Follow a live evaluation without installing anything
Included:MaterialsRecording€490
per person excl. VAT
1 and 2 Dec 2026
Two days, 20 teaching units (9am to 6pm)Marriott Leipzig or online
AI for chemists in research and development
Two days on your own questions: literature, structure proposals, evaluation, documentation. You leave with a workflow, not a demo.
Method
- Prompting, context, limits, checking results
- Documenting literature work and structure proposals traceably
Hands-on with tools
- Sketcher and MCP on your own molecules
- Your own use case on day two, supervised
Included:MaterialsCertificate30-day follow-up€1,490
per person excl. VAT
8 and 9 Dec 2026
Two days online, 20 teaching units (9am to 6pm)online
Am I allowed to? AI, data protection and data sovereignty in the lab
The questions that stall every rollout: where does my data sit, what may go to the cloud, when do I need my own model, and how do I make lab data usable at all.
Method
- GDPR, processing agreements, trade secrets, EU AI Act
- Cloud, EU hosting or local model: justifying the decision
- Making lab data usable: formats, metadata, legacy records
Hands-on with tools
- The same evaluation hosted and self-hosted, side by side
- Bookable for QA, IT and management together
Included:MaterialsCertificate30-day follow-up€1,490
per person excl. VAT
19 and 20 Jan 2027
Two-day intensive workshop (9am to 6pm)Marriott Leipzig
Structure elucidation with AI: NMR, MS, IR and UV/Vis
Spectra interpretation, purity assessment and impurity identification as tool calls. The focus is on when you may trust a proposal and when you may not.
Method
- Automatic assignment, confidence, applicability domain
- When a structure verification holds up and when it is guessing
Hands-on with tools
- CovaNMR, CovaMS, CovaIR and CovaUV on your own spectra
- Auto-annotation templates and purity workflows
Included:MaterialsCateringCertificate30-day follow-up€1,790
per person excl. VAT
28 Jan 2027
Half day online (9am to 1pm)online
AI readiness workshop: what works, what does not, what it costs
The entry point, no prior knowledge needed. On real spectra and chromatograms instead of text snippets, with an honest answer on where AI still gets things wrong.
Method
- How language models work and why they fail in chemistry
- How to tell a sound answer from an invented one
- Which lab tasks are worth it today and which are not
Hands-on with tools
- Follow a live evaluation without installing anything
Included:MaterialsRecording€490
per person excl. VAT
9 and 10 Feb 2027
Two days, 20 teaching units (9am to 6pm)Marriott Leipzig or online
Stability under ICH Q1: Arrhenius, shelf life and OOS/OOT
Turning multi-temperature data into a defensible shelf-life estimate, justifying model selection, and recognising when extrapolation cannot replace the real-time study.
Method
- Kinetics, Arrhenius, model selection by AIC, poolability
- OOS and OOT detection, significant change, residual diagnostics
Hands-on with tools
- CovaStab on your own or supplied data sets
- Documenting the limits of extrapolation properly
Included:MaterialsCertificate30-day follow-up€1,490
per person excl. VAT
23 and 24 Feb 2027
Two-day intensive workshop (9am to 6pm)Marriott Leipzig
ICH M7 in silico: mutagenic impurities and expert review
Structural alert triage, class 1 to 5 classification, and handling two complementary systems that disagree. With a traceable audit trail.
Method
- (Q)SAR, applicability domain, expert review under ICH M7
- Resolving contradictory predictions from two systems with justification
Hands-on with tools
- CovaTox, covatox_ich_m7 and batch assessment
- First-pass report templates (FDA, EMA)
Included:MaterialsCateringCertificate90-day follow-up€1,790
per person excl. VAT
9 and 10 Mar 2027
Two-day advanced workshop (9am to 6pm)Marriott Leipzig
GxP and self-hosted: CovaSyn MCP in regulated environments
How deterministic MCP tool calls support GxP requirements, how ML models are validated to inspection standard, and how CovaSyn runs as a container in your infrastructure.
Method
- Audit trail, ALCOA+, Annex 11, reproducibility
- Validating ML models: model quality, drift, applicability domain, explainability
Hands-on with tools
- Self-hosted deployment, signed container, EU hosting
- GAMP 5 validation templates
Prerequisite: Experience with sketcher and MCP
Included:MaterialsCateringCertificate60-day follow-up€1,990
per person excl. VAT
Planning for a team?
For two days, an in-house date costs less than individual seats from 7 people in one organisation, for a half day from 8. On your own use case and on a date of your choice.
A free way in before a workshop: the CovaSyn Open Lab, 45 minutes live on Microsoft Teams once a month.
FAQ
Frequently asked questions
Who are the workshops for?
For teams in chemistry, pharma and biotech, from the first contact with AI to the question of what is allowed in regulated environments.
Are there in-house dates?
Yes. You book in-house formats for your team. Formats and prices are in the catalogue.
How do I book?
Online, right in the catalogue. After payment you receive a confirmation with a calendar entry.
How reliable are the results?
14 to 41% without CovaSyn, 76 to 92% with CovaSyn. Measured on MolecularIQ, a benchmark we did not design (arXiv:2601.15279, Klambauer Lab, Institute for Machine Learning, JKU Linz). Every result is a first draft that your experts sign off.
When is an in-house date worth it?
For two days, an in-house date costs less than individual seats from 7 people in one organisation, for a half day from 8. On your own use case and on a date of your choice.
What if no open date suits us?
New open dates appear in the catalogue as soon as they are set. An in-house date for your team can be planned independently of that, on a date of your choice.
Next step
Give your team a shared starting point.
Open dates and in-house formats at a fixed price, bookable online.