Ask a formulation question and DeepC runs the study it calls for. You get a cited answer and, when the work needs one, a formulation, an FTO opinion or a regulatory report written to its framework.
DeepC matches the work to the question, from a cited answer in a sentence to a full report.
01
DeepC writes the brief and initiates the study the question calls for, under the regulatory basis it follows where one applies.
02
Each study moves through the stages it declares, shown as it enters them. It goes back to an earlier stage when the evidence sends it there, and skips one that does not apply.
03
Every source the study consults is counted, stage by stage and by kind: regulations, patents, toxicology, the literature and your own documents.
04
The deliverable streams into the conversation in the structure its framework expects, every claim cited.
05
A finished study is filed under a number from your organisation’s own sequence for that study, carried into the library and onto the exported report.
Formulation science is visual, so DeepC answers in kind: structures, charts, figures from the documents and the deliverables themselves, where they carry information.


Name a compound by SMILES, InChI, CAS number or name and DeepC draws it, highlighting the functional group the question turns on.
When a dissolution plot or a structure in a regulatory document settles the question, DeepC shows it with its page and source.
Drawings, process diagrams, data figures, calculators whose settings become your next question, and dashboards that combine them.
Structured work opens beside the chat and streams in as it is written.
Any answer or whole conversation exports; analytics and regulatory reports compile with LaTeX typography. A study’s report carries its record number and closes with its procedure record: each stage, its time and the evidence behind it.
Each job has its study, and each hands back a deliverable your team can take forward.
01
Every design opens with a QTPP per ICH Q8(R2), then three to five candidates, each from a different angle and ranked against the others.
02
A best-estimate Q1/Q2 composition, graded for confidence, with the dissolution target, the patents around it and where to source each ingredient.
03
Variants that name what each gives up for what it gains, or an Optimize campaign that learns from your bench results.
04
Approved-product landscapes, sequence-similarity FTO for antibodies, and formulation candidates for mAbs, bispecifics, ADCs, LNP-mRNA, AAV and cell therapies.
05
ERA, extractables, leachables, nitrosamines, OEL and PDE, each in the structure its framework prescribes, with a Klimisch annex and the sign-off left to your QP.
06
An FTO opinion with claim charts and design-arounds, then a patent application template for your attorney.
07
Upload a dataset for statistics and figures in the conversation, or a full report from a data analysis.
We benchmarked DeepC against our external toxicology consultancy on a substance-specific risk assessment. DeepC delivered a report our QP rated equal or stronger in around five minutes, versus the roughly one month our incumbent service took.
Everything a research team needs around the conversation, from projects to a memory of how you work.
Any file type or a whole folder, up to 5 GB a file. PDFs and Office documents are parsed so DeepC can search and cite them.
Upload a dataset and DeepC runs the statistics and plots in the conversation, or runs a data analysis for a formal report.
Conversations, files and analyses together, with Instructions for every conversation and Knowledge DeepC cites. Owner, Editor and Viewer roles.
A small set of notes about you, your projects and your feedback, which you can inspect, edit or switch off.
Every report, analysis and campaign, by month and by record number; your organisation’s documents and connected Microsoft 365 libraries.
Search every chat, point back to earlier work, keep typing while an answer streams, dictate, and keep a temporary chat out of your history.
Share a conversation by read-only link for 30 days, and comment on any finding.
The studies sit in three suites, one for each kind of work: formulation, IP and research; biologics; and regulatory safety.
Research suite
Formulation, deformulation, optimisation, FTO, patent drafting, VCM process design, literature review and data analysis.
Biologics suite
Biologics research, FTO, formulation and CMC. Antibodies, ADCs, bispecifics, mRNA-LNP, AAV, cell therapy.
Safety suite
ERA, extractables, leachables, nitrosamines, OEL and PDE, returned as dossier-ready output.
Each report follows the structure of the text it is filed against, down to the clause.
From a first call to a pilot on your own programme.
A call with a formulation lead, who is in touch within a business day.
On your own programme, scoped with your team.
The AI model providers behind DeepC retain none of your content and train nothing on it.
Prefer email? info@deepceutix.com