Answers to the questions we hear most.

SpeciGen is the AI-native biospecimen operations platform for clinical trials, using agentic AI to accelerate study setup, continuous cross-vendor reconciliation, operational oversight, and submission-ready eTMF closeout. Sponsors, CROs, and academic PIs use it standalone or across their existing clinical stack. AI accelerates the work, humans retain control, and a deterministic core maintains the study record. 21 CFR Part 11 attested.

Yes — both, and they mean different things. AI-native means large language models are first-class participants in the platform, not a chatbot bolted onto a legacy configuration engine. Study setup, reconciliation reasoning, anomaly clustering, and natural-language query all sit on the same LLM plumbing shared across dev and production. Agentic AI means SpeciGen ships agents — AI systems that take a goal ("configure this study from these documents"), plan the steps, act using platform tools, and iterate based on human feedback. The Agentic Study Setup is our first shipped agent; more are on the roadmap. Underneath both: every state SpeciGen records — a subject, a visit, an expected specimen, a signed discrepancy — is written by deterministic code against schema-validated data, signed by a human. AI proposes; humans approve; the deterministic core records. That combination — AI-native by architecture, agentic by design, deterministic where audit matters — is the moat.

SpeciGen is built for clinical trial sponsors, contract research organizations (CROs), central labs, and biotech / pharma companies running Phase 1–4 studies with rich PK, ADA, biomarker, or pathology workflows — including biosimilars and oncology programs.

Yes. SpeciGen's 21 CFR Part 11 compliance has been attested by an independent third-party assessor in 2026-Q2. The platform implements electronic records with a hash-chain audit trail, SHA-256-bound electronic signatures, role-based access, and unique user identification via Amazon Cognito. Attestation report available under NDA — see Trust & Security.

SpeciGen is priced per-study, per-year — flat, regardless of site count, patient count, vendor count, or feed count. A typical 3–4 year study lands around USD 70,000 total, with implementation, protocol amendments, and ongoing support all included. A 3-month evaluation pilot is also available at nominal cost.

Yes. SpeciGen ingests data from any EDC, LIMS, central lab, or biomarker vendor via CSV, Excel, SAS XPT, PDF, pathology image, or REST API. Native mapping templates exist for Medidata Rave, Veeva Vault, and CDISC SDTM domains; vendor-specific mappings are configured during study build.

SpeciGen is hosted on Amazon Web Services (US region by default; EU regional deployment available for in-country residency requirements). All data is encrypted at rest with AES-256 (KMS-managed keys) and in transit with TLS 1.3. Identity is managed via Amazon Cognito with MFA-enabled access for administrative roles. Full detail on Trust & Security.

Yes. SpeciGen ships a dedicated Site Coordinator workspace that runs on desktop and as a mobile progressive web app (add to home screen on iOS Safari or Android Chrome). Coordinators can log a site visit, generate or upload electronic requisitions, print sample labels, scan kit barcodes with the phone camera, and edit raw sample fields — all scoped to their site. Two role variants exist: BLINDED (no cohort, arm, or lab results visible) and UNBLINDED. Every change is audit-logged with a before/after diff.

Yes. SpeciGen mints sample barcodes using a per-study format with an atomic sequence generator, renders real Code128 graphics on the label PDFs, and produces sample-anchored electronic requisitions (eReqs) either automatically on sample creation or on demand. Bulk label printing is available for site coordinators, and eReq PDFs carry full version history per sample.

Yes — as an optional service. Sites that don't have a label printer, or sponsors that prefer to receive pre-printed kits with matched barcodes, can have SpeciGen print the label sheets on our stock and ship to each site directly. Priced at cost, quoted per study. The self-serve PDF export in the app remains available for sites that print in-house.

A central lab’s portal only knows what its own lab handles. Most modern trials involve more than one lab — a central lab, a PK/ADA lab, a biomarker lab, sometimes a specialty genomics or pathology partner — plus EDC data from the sites, kit shipments, and site-level collection events. No single lab is naturally the neutral system of record across all of that. SpeciGen is the vendor-neutral layer that reconciles what each lab reports against what the protocol expected, and gives the sponsor a single operational view they can trust regardless of which vendor is involved. Central lab portals remain a first-class data source into SpeciGen; SpeciGen sits above them, not in place of them.

Reconciliation is one output. What SpeciGen actually maintains is expected-vs-observed state at every node of the specimen journey: protocol → amendment → subject → visit → expected specimen → collection → processing → aliquots → shipment → lab receipt → assay → result → storage. Because SpeciGen holds the projected schedule from the protocol AND the observed events from every vendor, the platform knows — in real time — whether every expected biospecimen actually exists, is where it should be, and reached the assay it was intended for. Reconciliation is the report that falls out of that model. Anomaly detection, sponsor status reports, discrepancy trackers, and eTMF packages all reuse the same underlying graph.

A new study is typically live in 2–4 weeks from kickoff, and much of that time is human review rather than translation work. SpeciGen’s Agentic Study Setup (our first shipped agentic-AI capability) reads your protocol PDF, Schedule of Activities, and lab manual and produces a draft study configuration — cohorts, arms, visit schedules, visit windows, and specimen collections — where every extracted value cites its source document and page. Your team accepts, edits, or rejects each proposal. No custom engineering: study metadata, vendor bindings, and column mappings are then finished in the SpeciGen UI. Native mapping templates exist for Medidata Rave, Veeva Vault, and CDISC SDTM domains; custom vendor feeds use SpeciGen’s column-mapping tool, which the sponsor’s data manager or CRO can operate directly without SpeciGen engineers. Implementation, protocol amendments, and ongoing support are included in the flat per-study price.

Yes. Amendments run through the same Agentic Study Setup as the initial configuration, but instead of proposing a fresh study the agent proposes a diff against the current configuration — a visit added, a cohort renamed, a specimen collection changed. Your team reviews and approves each line. Earlier collections keep the schedule they were collected under so historical audit is preserved; future visits pick up the new expected specimens; reconciliation just works. Amendments are handled as part of the flat per-study price — no professional-services line item, no reimplementation, no data lock-in.

SKIA — SpeciGen Knows It All (formerly Ask SpeciGen) — is the natural-language query interface built into SpeciGen. Users ask questions in plain English (“which baby stool samples have been collected but not shipped?”) and get back the answer as a table, count, chart, or narrative summary, with CSV or XLSX export. SKIA queries the master sample tracker, raw sample data, projected schedule, bioanalytical results, kit inventory, discrepancy log, and the event lifecycle ledger.