Skip to content

Platform

Architecture

One principle governs the whole system: deterministic NLP establishes what documents say; LLMs reason only over verified evidence. Everything else follows.

Business view — the pipeline

  1. 1

    Ingestion & classification

    16 formats normalized into one canonical view; rule-based classifier assigns document type.

  2. 2

    Privacy boundary

    Fail-closed de-identification before anything downstream executes.

  3. 3

    Clinical understanding

    Deterministic NLP: entities, assertions, temporality, coreference, relations.

  4. 4

    Terminology linking

    Mentions map to SNOMED CT, RxNorm, LOINC, ICD, HPO, MedDRA, HGVS concepts.

  5. 5

    Biomedical IR

    Every fact lands in one canonical contract — provenance-stamped, assertion-qualified, time-anchored.

  6. 6

    Derived structures

    Timeline builder, provenance-carrying evidence graph, hybrid retrieval index.

  7. 7

    Evidence packs

    Retrieval assembles cited evidence neighborhoods for each question.

  8. 8

    Reasoning firewall

    Claims validated against evidence (token overlap + numeric fidelity); unsupported statements dropped.

  9. 9

    Audited intelligence API

    Answers ship with citations, claim verdicts, uncertainty notes — and usage meters fire exactly once.

Explore each stage interactively in the intelligence pipeline explorer.

Technical view — AWS deployment

Compute

Amazon EKS · separate node groups for API pods, CPU-bound NLP workers, GPU inference pool (vLLM serving local Qwen/Llama/Mistral models).

Relational & graph data

Aurora PostgreSQL with Apache AGE for the evidence graph (openCypher internally — never exposed publicly).

Search

Amazon OpenSearch for production-scale hybrid retrieval (BM25 + vector-capable indexes).

Object storage

S3 with SSE-KMS encryption, versioning, and lifecycle policies mapping logical storage classes.

Async & events

Amazon MSK / EventBridge carry immutable usage events and job orchestration between services.

Secrets & keys

AWS KMS customer-managed keys; Secrets Manager holds privacy secrets, provider keys never present in application code.

Edge

Route 53 → WAF → ALB → private pods. No public IPs inside the cluster; EKS endpoint private.

Capability abstraction: public APIs expose what the platform does — never which engine served it. A CI gate scans every payload for infrastructure terms so engine tiers, routing rationale, and node traces stay internal. Clients build against stable capability contracts with versioned semantics.