Giovanni Di Mauro, PhD

Small and mid biotechs hit the same wall around clinical decision moments: pre-launch positioning, lifecycle re-reads, BD diligence, scenario-planning for a competitor readout. The internal headcount is not there. The big consultancies sell six-week engagements at six figures, with junior analysts doing the synthesis on top of recycled databases. The boutique firms hide pricing and methodology.

Delvant is the answer I wanted to exist when I was reading those reports from the buyer side: fixed scope, fixed fee, audit-format methodology, every claim tagged with its evidence tier, every source locked to a primary registry, every report personally signed by the person who built it.

Doctoral work in the Maulide group (Institute of Organic Chemistry, University of Vienna) centred on mechanistic organic synthesis: designing and validating new bond-forming reactions, isolating intermediates, building mechanistic proposals from kinetic and spectroscopic evidence, and defending them against alternative mechanisms. That work taught a habit that turns up everywhere in Delvant deliverables: any analytical step that cannot be defended against an alternative mechanism is flagged, not buried.

Alongside the synthetic work, the PhD included an in-silico drug-design strand: structure-based ligand design, conformational search, DFT calculations on small-molecule reactive intermediates and protein-ligand modelling for series prioritisation. The point of that strand was not to publish a virtual-screening pipeline; it was to put a quantitative layer next to a wet-lab hypothesis and force the two to either agree or expose where the model was too coarse. The same posture sits inside Delvant's evidence-tier system. A claim that survives only because the model is too coarse is not a Sourced claim; it is a Hedged claim, and it ships with that label.

A research visit at the Houk lab at UCLA extended the computational side: transition-state modelling of organic mechanisms under K. N. Houk's group, applying DFT and post-Hartree-Fock methods to mechanism questions raised by wet-lab work in Vienna. No publication came out of that visit. The methodological habit did: build a quantitative model, state where it can be falsified, and accept that a model that cannot disagree with reality is not a model. That habit is the same one a buyer should expect from a Delvant analytical step, regardless of therapeutic area.

None of this training is clinical-affairs training. Delvant is explicit about that: it is not a clinician, not a regulatory affairs firm, not a CRO. What this training does provide is the ability to read primary biomedical literature at experimental depth, distinguish a real comparative claim from an apparent comparative claim, and recognise when a published trial conclusion outruns its actual evidence base. That ability is the core engine of the deliverable, and it is what justifies the per-claim evidence-tier discipline rather than the trust-me-it-is-good-research posture sold by larger vendors.

Delvant is one person and a documented methodology. That is the point, not a limitation. There is no account manager rotation, no junior analyst handoff, no agency-side scope creep. The person who reads your question is the person who reads the papers, builds the model, writes the PDF, and signs it.

If the question is outside Delvant's tier coverage, I will say so on the first call. If a Phase III readout has just landed and changes the framing, I will tell you before scoping. The published methodology and sample report are deliberately open so the buyer can verify both before any contract.