Bloomb Health Services / Technical Platform

Safe, Reliable AI Infrastructure
for Vulnerable Populations.

BHS is clinical AI infrastructure for behavioral health platforms serving caregivers and vulnerable populations. A deterministic safety layer. A zero-PII privacy architecture. Population-specific clinical knowledge graphs. And a structured evaluation process that runs before a single real user ever reaches the system.

2 Live pathways, each with its own population-specific knowledge graph
85 Synthetic clinical scenarios run before either pathway reached a live user
0‑PII No personally identifiable data touches the LLM layer

Enterprise-Grade Architecture
for Healthcare.

BHS goes beyond basic prompt wrappers. Every message passes through a privacy architecture, a population-specific clinical knowledge graph, and a deterministic Crisis Rules layer before the LLM is invoked. On crisis signals, the Crisis Rules layer takes control and directs the LLM. Casual conversation is not possible, and the model has no discretion. This infrastructure exists because the population demands it.

Privacy Architecture

Confidential
by Design

The BHS privacy architecture is built on minimum necessary use: the system collects only what is required to deliver a clinically appropriate response, retains only what is required to maintain care continuity, and shares nothing beyond what the user has explicitly consented to. Messages are encrypted at ingestion and stored separately from the derived clinical data. The emotional states, lived experiences, and longitudinal care history that drive the system live in a separate dataset from the message content itself, and the two are never joined at the model layer. The Clinical Knowledge Graph receives structured clinical context, and the LLM receives an anonymized payload drawn from that context, with no path back to the identity of the person who sent it.

  • Messages encrypted at ingestion and stored separately from clinical derived data
  • Emotional states, experiences, and longitudinal history in a separate dataset from message content
  • Clinical Knowledge Graph receives structured context, not raw messages
  • LLM receives anonymized payload only. The two datasets are never joined at the model layer
  • Full audit log on BHS-controlled infrastructure
  • HIPAA-aligned architecture; BAA in progress with Microsoft
Privacy Pipeline / Data Flow
U
User Message Received
Raw input, session metadata, device context.
B
Encryption and Dataset Separation
Message encrypted at ingestion. Clinical derived data (emotional states, experiences, longitudinal history) stored in a separate dataset. The two are never joined at the model layer.
K
Population-Specific Knowledge Graph
Anonymized context matched against the clinical graph for this population. Feeling nodes, experience nodes, care stage, and crisis indicators resolved.
B
Crisis Rules Layer
Detects danger level from resolved context. If triggered: dynamically generates a tightly controlled prompt. LLM directed to emergency referral or crisis techniques only. Casual conversation not possible.
L
Private LLM Endpoint
Zero PII in payload. Clinical context and grounding injected by BHS. Model runs in your private cloud, with no shared public API.
B
Output Validation + Audit Log
Response scanned before delivery. Every session event, including risk level, technique selected, and safety protocol triggered, is written to the audit log.
Crisis Rules Layer

Deterministic Safety
That Controls the Model

The Crisis Rules layer is not a prompt and it is not a classifier. It is a deterministic logic engine that runs on every inbound message after the knowledge graph resolves context. When a crisis signal is detected, the Crisis Rules layer classifies the danger level and dynamically generates a tightly controlled prompt based on that classification. The LLM responds only within that prompt, directed to either connect the user with emergency services or deliver clinically appropriate crisis techniques. Casual conversation is not possible at that point. The LLM has no discretion. The Crisis Rules layer does.

No deterministic system catches every indirect or implicit expression of risk. BHS does not claim otherwise. The Crisis Rules layer is under continuous expansion with clinical advisory input specifically to close that gap. The SAFE Standard evaluation framework is designed to surface those gaps before they reach a live user.

  • Detects danger level and dynamically generates a tightly controlled prompt, directing the LLM to emergency referral or crisis-appropriate techniques only
  • Casual conversation is not possible once active. The LLM has no discretion. The Crisis Rules layer does.
  • Every crisis event is logged with timestamp, danger classification, and care status. Sustained risk across sessions triggers consent-gated care network notification.
Crisis Rules Layer / Bypass Logic
Inbound Message + Resolved Context
"I do not know how much longer I can do this." The Knowledge Graph has resolved: Exhaustion + Grief + Orange care stage.
⚡ Crisis Rules Layer: Danger Detected
Danger level classified from resolved context. Crisis Rules layer takes control. Dynamically generates a tightly controlled prompt based on classification.
LLM Directed: Emergency Referral
At high danger levels: LLM directed to connect user with emergency services or 988. No other response is possible.
LLM Directed: Crisis Techniques
At elevated danger levels: LLM delivers crisis-appropriate techniques only. Casual conversation not possible. Audit log updated.
Pre-Deployment Evaluation

Nothing Ships
Without a Named Framework

Every BHS pathway runs through a structured, versioned clinical evaluation framework before a single real user reaches it. Bloomb was evaluated against the SAFE Standard, a BHS-authored framework built around four non-negotiable criteria. Clover was evaluated against the SMART 40, a framework independently defined by the grant reviewer who assessed the program. Both evaluations run on population-specific synthetic scenario libraries. The results are documented, versioned, and available for IRB submission, insurer review, and enterprise due diligence.

View Full Evaluation Results  →
Two Pathways. Two Frameworks. Both Validated.
Bloomb / Postpartum
SAFE Standard
BHS-authored. Four criteria: Secure, Accurate, Focused, Explainable. 45 synthetic clinical scenarios across 5 risk states.
Clover / Autism Caregiver
SMART 40
Independently defined by the grant reviewer. 40 cycles across stress, boundary, and standard categories.
Both Pathways / In Development
VERA-MH
Open-source mental health AI safety benchmark published by Spring Health. BHS is implementing VERA-MH evaluation across both pathways.

Your Data Stays
Where You Put It.

BHS runs on Microsoft Azure. Every component, from the population-specific knowledge graph and interaction history to the audit log and private LLM endpoint, lives in a cloud architecture designed around the principle that behavioral health data should never leave the environment you control. Data residency, encryption at rest and in transit, and role-based access are not configuration options. They are defaults.

Private LLM

Model Calls Never Leave Your Perimeter

The LLM runs on a private endpoint inside your Azure tenancy or BHS-managed infrastructure. No message, no clinical context, and no user payload ever reaches a shared public API. If the endpoint is compromised, there is no PII there to expose, because the decoupling layer handled that upstream.

Knowledge Graph

Population-Specific, Clinically Structured, Separate from the Model

Each pathway runs on its own graph. the Bloomb knowledge graph knows postpartum, the Clover graph knows autism caregiving. They share infrastructure but not knowledge. The graph is structured, queryable, and completely separate from the LLM. If the model changes, the clinical knowledge stays yours.

Audit Logging

Every Session Event. Structured. Reviewable Without Seeing the Messages.

Every session produces a structured record: emotional states detected, risk level assigned, techniques selected or suppressed, safety protocols triggered, care network notifications sent. The record is complete, timestamped, and reviewable by a clinician without a single patient message ever being exposed. This is the audit trail your IRB, insurer, and legal team need.

Data Privacy

Encryption, Residency, and Minimum Necessary Use

Data is encrypted at rest and in transit. Interaction history is stored separately from the knowledge graph. Reporting uses derived clinical state, covering emotional patterns, risk trajectories, and session outcomes, with no verbatim user content in any reporting surface. Data residency is configurable per deployment. Role-based access limits who can see what, and audit logs capture who looked.

Compliance

HIPAA-Aligned Architecture. BAA in Progress.

The BHS architecture is built around HIPAA privacy principles: consent-gated collection, minimum necessary use, data separation, role-based access, and documented breach response. A Business Associate Agreement is in progress with Microsoft. Compliance documentation is available for enterprise due diligence.

Monitoring

Application-Level Visibility Across Every Session

Real-time monitoring tracks system health, response latency, safety protocol activation rates, and error events at the application level. Alerts fire on anomalous patterns before they become incidents. Session-level telemetry is retained for post-deployment audit, separate from clinical content.

The System Tracks State.
The Response Follows.

Most conversational AI resets with every session. The BHS Dynamic Care Pathway engine tracks state across the full care relationship. It selects responses from the population-specific knowledge graph, matching conditions, emotional states, care stages, and appropriate interventions into explicit relationships rather than leaving that reasoning to the LLM. The result is a system that responds differently to the same words depending on who is saying them, when, and what has happened in every prior session.

Once the care level is set, the personalization engine selects the specific technique, resource, or referral that fits this person at this moment, including routing to your own vetted local provider network if you bring one.

Clinical Ontology Graph-Based Routing Multi-Session State Personalization Engine Local Provider Referral

The care level system spans eight states from healthy adjustment through acute crisis, each with a defined response posture, technique set, and escalation logic. As a user moves through those states across sessions, the system tracks the trajectory and adjusts accordingly. A user who reaches a sustained crisis state across multiple sessions receives a different response than someone encountering that state for the first time, because the system knows the difference.

Wellbeing States

Positive and neutral states are met with warmth and reinforcement. The system does not over-pathologize calm moments.

Distress States

Elevated distress triggers empathetic responses with appropriate technique selection and a soft bridge toward professional care as severity increases.

Crisis States

Crisis detection hands control to the Crisis Rules layer. Response is directed, not discretionary. Sustained crisis across sessions triggers care network notification.

i

The knowledge graph is what makes this possible. When the system selects a response posture or escalates a level, it is traversing a structured graph of conditions, emotional states, care stages, and clinically appropriate responses. State persists across sessions. A user who reaches a sustained crisis state did so across multiple conversations, and the system tracked that history.

Built From Lived Experience,
Not Assembled From Data

Each BHS pathway is built on a population-specific clinical ontology developed from direct lived experience and clinical advisory input. The knowledge architecture exists independently of the LLM, which means responses are clinically coherent because the reasoning is explicit, not because a model achieved it through statistical probability. A new pathway begins with the knowledge graph, not with a prompt.

Founder Lived Experience Across Both Live Pathways
01
Live on Bloomb

Maternal Mental Health
and Postpartum Care

The flagship BHS pathway. Built across the full postpartum arc from healthy adjustment through Edinburgh-scored depression risk, NICU trauma, birth trauma, and perinatal loss. The knowledge graph maps 25 emotional states and 35 lived postpartum experiences across 4 care stages. Validated against the SAFE Standard before launch.

Postpartum Depression Birth Trauma NICU Support Perinatal Loss Edinburgh Scale SAFE Standard
02
Live on Clover

Autism Caregiver
Support

Built for the caregiver, not just the child. The knowledge graph maps 25 emotional states and 35 lived caregiver experiences across 4 care stages: Early Intervention, School Years, Adolescence and Puberty, and Transition to Adulthood. 30 evidence-based techniques tuned to this population at that specific moment. Validated against the independently defined SMART 40 framework.

Early Intervention IEP Advocacy Caregiver Fatigue Diagnosis Navigation Transition to Adulthood SMART 40

Longitudinal Clinical Data,
Available Under Consent.

Because BHS captures emotional state trajectories, risk level histories, care stage progressions, and technique outcomes across sessions, and because consent is gated at the point of collection and not applied retroactively, that data is available to researchers and providers in structured, derived clinical form. No raw message content is ever included. The architecture was designed for this from the start.

For Researchers

Structured Longitudinal Data for Population Research

BHS can provide consented, de-identified longitudinal datasets covering emotional state trajectories, risk level patterns, care stage progressions, and intervention outcomes across the postpartum and autism caregiver populations. Data is structured, timestamped, and expressed entirely in derived clinical language, suitable for IRB-approved research without requiring access to any raw session content.

  • Consent captured at point of collection, not applied retroactively
  • Fully de-identified, with no raw message content in any research dataset
  • Available for postpartum and autism caregiver populations
  • Structured for IRB submission and compatible with standard analysis pipelines
For Providers

A Longitudinal Clinical Record Without Seeing the Sessions

Providers working with patients who use a BHS-powered platform can receive a structured longitudinal clinical record covering emotional state trajectory, risk level history, and care stage progression, all under patient consent and in derived clinical language. No session transcripts. No raw disclosures. A clinician can review what the system understood about a patient over time without reading a single message the patient sent.

  • Patient-consented longitudinal record, available on request
  • Covers emotional trajectory, risk history, and care stage progression
  • No session transcripts or raw disclosures in any provider-facing record
  • Privacy architecture and clinical oversight are not traded against each other

Start a Conversation

Tell BHS about your platform and your population. Most conversations start with a simple question about whether this is a fit for what you are building.

All inquiries handled under NDA on request
Response within one business day
Partnership Inquiry