Hridhaan Systems

Evidence-grade AI for regulated decisions

We read research and development documents, map them against the rules that govern them, and return citation-backed, auditable output. The decision stays with the expert. The evidence trail is produced for them.

evidence pack · assessment record rule pack 01 · dual-use screening
Findings — each traced to source
Criterion touched — enhancement of transmissibility
“…serial passage in a permissive cell line to select for variants with increased receptor affinity…”
§ Proposed work, p. 14, ¶3 · document v2 · 2026-08-11
Criterion touched — pathogen of pandemic potential
“…the isolate is handled under BSL-3 containment for the duration of the challenge study…”
§ Methods, p. 22, ¶1 · document v2 · 2026-08-11
Audit record
Framework versionUSG DURC/PEPP 2024-05
Rule packdual-use v1.4.2
Model & promptpinned · hash 8f2c1a
ConfidenceReview required
Routed tonamed reviewer
Assessed2026-08-11 09:41 UTC
Reproduciblere-runnable on demand

Illustrative output. Every finding cites the passage that triggered it, against a dated version of the rulebook.

Frameworks the engine is pointed at
US DURC & PEPP criteria EU Regulation 2021/821 Annex I Australia Group lists EU Biotech Act 21 CFR Part 11 GVP
01 — The problem

Judgement at a volume no expert team can cover — and proof of how each one was reached

Regulated organisations are asked to assess documents faster than their experts can read them, then to reconstruct the reasoning years later for someone who was not in the room.

A funder

Several hundred proposals a cycle

Must confirm that none of them creates a biosecurity risk — before money is committed.

A pharmaceutical company

Every collaboration and exchange

Must screen technical content against export control lists, and prove which list version applied.

A quality team

Every deviation, every investigation

Must reconstruct its reasoning years later, for an inspector who arrives without notice.

All three are the same shape of work: read unstructured documents, apply an external rulebook, record the reasoning, decide.

Today this is done by hand. The cost is not the reading — it is the evidence. Guidance issued to research funders by the Nuclear Threat Initiative instructs them to limit biosecurity review to shortlisted proposals in order to conserve resources. The organisations that need this most have already written down that they cannot afford to do it properly.

General assistants have made the reading cheap. They have not made the evidence admissible.

02 — The engine

One pipeline. Five rule packs.

The engine does not change between use cases — only the rulebook it is pointed at, and the dated version of that rulebook in force at the moment of assessment.

01

Read

Proposals, protocols, SOPs, technical files and submissions — in full, not titles and abstracts.

02

Map

Against an external framework, at a specific dated version of that framework.

03

Cite

Structured findings, each one quoting the exact source passage it came from.

04

Record

A complete audit trail: what was assessed, against which rule version, by whom, when.

05

Route

To a named human reviewer, who makes the decision and signs it.

The output is not an answer. The output is a decision-ready evidence pack.

Findings, citations, confidence, and the documentation a reviewer or an inspector will ask for.

03 — Rule packs

Five rule packs, in the order we intend to build them

Same engine. Different rulebook. Each one produces the artifact its buyer already has to write by hand.

04 — Deployment

It runs inside your tenant. Documents never leave.

Rule packs and logic arrive as signed, versioned bundles over an outbound connection — the same pattern any licensed enterprise software uses. Where a customer permits no outbound connection at all, bundles are delivered as signed offline packages.

Residency

Your environment

Confidential proposals, patient data and trade secrets stay where they already are.

Updates

Signed, versioned bundles

You choose when a rule pack version changes, and every assessment records which one it ran against.

Air-gapped

Offline delivery

Where no outbound connection is permitted, the same bundle is delivered as a signed offline package.

This matters commercially as much as technically: it removes the largest procurement objection before it is raised, and it keeps a single codebase across every customer.

05 — Durability

What holds when the models improve

Intelligence gets cheaper with every model release. Accountability does not. Four things survive the next frontier model, and they are the product.

01

Versioned determinism

A regulated workflow needs the same input to return the same output three years later, or a documented reason why not. That guarantee cannot be given by a vendor whose business depends on continuous upgrade.

02

Validation and named accountability

Someone must sign that the system is validated for its intended use. The foundation model vendors will not sign. Transferring that accountability is the product.

03

Curated rule packs and expert-labelled evaluation sets

The model knows the world. It does not know this organisation's interpretation of Annex I, or its SOP hierarchy. That is curation work — labour, not intelligence — and model releases do not erode labour.

04

Write-back into the system of record

A general assistant reads. It does not write into a safety database or quality system under change control, with e-signatures and an audit trail that satisfies 21 CFR Part 11.

Where the investment goes

  • Versioning and determinism
  • Automatic generation of validation documentation
  • Expert-labelled evaluation sets
  • Part 11-compliant write-back into systems of record
  • Dated versioning of rule packs

Where it does not — free in the next release

  • Better prompts
  • Better retrieval
  • Multilingual capability
  • Summarisation quality
  • Raw model performance

The product is partly unglamorous plumbing — versioning, signatures, documentation, audit. That is the good news: model releases commoditise intelligence, not plumbing.

06 — Why now

Regulatory load is rising, not settling

Policy

Rewritten three times

US policy on high-risk life sciences research was rewritten in 2024, again in 2025, and again in July 2026. The EU Biotech Act extends duties from funders to companies and laboratories.

Incentives

Not the assistants' problem to solve

Closing the evidence gap works against the general assistants' own business model, because it would mean freezing model versions on a customer's behalf.

Category

Proven, and unoccupied

The adjacent export-control software market has sold list-based screening for twenty years — the category and the willingness to pay are established. No vendor was found operating at the document layer in English, German or French.

The window is real but not permanent. Established vendors can build this once the market is visibly large. The defence is to be inside a validated workflow before that happens.

Start here

A fixed-scope assessment engagement, on your own documents

Four weeks. Your documents, your frameworks, inside your environment. It produces measured baseline numbers and a working prototype. Success criteria and the route to production are agreed in writing before it starts.

Book a call

Or write to hello@hridhaan.ai.