Platform & handoff

Technical credibility — evidence, calibration, and claims.

The technical credibility surface: what evidence, calibration, and claims DiscoveryLab stands on.

Technical credibility

Designed for constrained discovery, not infinite generation.

01Condition-first target selection

02Receptor-conditioned generation

03BioScout mimicry from evolved peptide systems

04Structured candidate families, not single sequences

05v2 non-natural amino-acid support

06Genetic programming with bounded mutation/crossover

07Scientific claim-boundary ledger across every assay capability and result

08Public calibration import pipeline for ChEMBL, BindingDB, and PDBbind rows

09Protein-language-model task-head replacement plan for descriptor assays

10ESM-2 3B embeddings generated for all 221 repository candidates

11Boltz-1 / AWS HealthOmics complex-prediction request and contact-map preview

12Pareto-front and active-learning next-batch design

13Wet-lab prep package with 10 selected candidates, 13 synthesis specs, 10 assay protocols, and return template

14In-silico scoring gates at every stage

15Pathway mechanism verification: reachability, blockade, safety, conservation

16Lean 4 audit artifacts and external verification receipts

17PathwayLean reusable package plan with checksums and file list

18Perturbation evidence scoring against encoded assumptions

19RealityKit molecular review (macOS / visionOS)

20Five-workspace progress ribbon as the only app navigation

21Keychain-backed LabSpace notification token handling

22LabSpace wet-lab feedback loop

23Recursive Discovery Loop with improvement proposals, regression gates, approval receipts, and rollback

24Local tutor fallback plus Bedrock/Kendra grounding contract