K Health
Primary care / virtual physician intake. Two-agent pipeline: one extracts the relevant EMR subset, second answers grounded to that filtered history. Architecture explicitly diverges by patient condition (diabetic vs heart-failure get different responses to same intake).
At a glance
- Type
- EMR-filtered two-agent pipeline
- Tier
- T1
- Created
- 2016 (founded 2016 by Allon Bloch Ran Shaul et al.; New York)
- Latest release
- not applicable — not OSS
- License
- not applicable — not OSS
- GitHub
- not applicable — no GitHub repo
- Pricing
- searched not found
- Funding
- $50M $900M val Series F (approx) · 2024-07
Taxonomy
- storage
- relational
- retrieval
- similarity
- persistence
- long-term
- update
- extraction
- unit
- episode
- governance
- auditable
- conflict
- none
When to use
Optimised for: HIPAA compliance + clinical-grade provenance + EHR integration
Anti-fit: not for non-healthcare verticals; must operate under HIPAA / regional health regulation
Pros & cons
Pros
Long-running clinical AI with memory grounded in millions of de-identified patient-symptom trajectories — actual ground-truth corpus.
Cons
K Health's data is proprietary — memory benefits don't transfer to other products; primary care scope.
Claims & capabilities
Matched physician decisions in 2/3 of real cases. Fine-tuned on Gemma 3 via Vertex AI. Live on Mayo Clinic Platform (May 2025); Northwell partnership; HBS case study.
Technical surface
- API surface
- searched not found
- Backend storage
- searched not found
- Deployment
- searched not found
- Embedding model
- searched not found
- Multi-tenancy
- searched not found
- MCP
- no MCP support advertised — vertical product, no MCP server / client integration documented
- A2A
- no A2A protocol support advertised — vertical product, no A2A integration documented
- OpenTelemetry
- no OpenTelemetry integration advertised — vendor logs/observability not publicly documented
Similar systems
Other vertical / domain-specific ai memory in the catalog, ranked by inbound references.
- NVIDIA ReMEmbR T3
Builds long-horizon memory by captioning video segments with VILA, storing captions with timestamps + 3D position coordinates in MilvusDB. At query time, LLM iterates retrieval across text, time, and position modalities. Deployed on Nova Carter robot (Jetson Orin).
- Abridge T1
Clinician-assist ambient documentation. Source mapping: every AI-generated summary element traced back to the source utterance. Audit-and-trust layer over episodic memory. Built on proprietary 1.5M+ medical-encounter dataset.
- ASAPP GenerativeAgent T1
Treats memory as first-class architecture. Captures the digital footprint of every interaction; retrieves preference and history at engagement time. Public example: airline knowing a frequent flyer wants aisle seats with her son — preference-aware, not just history-lookup.
- BenevolentAI T1
Target identification / drug repurposing / mechanism tracing. 85+ data sources, petabyte-scale, rebuilt every few weeks. Wet-lab results re-enter the graph and shift downstream predictions — institutional experimental memory closing a feedback loop.
- Causaly T1
Drug discovery / target identification / causal mechanism tracing. The graph is the memory: 7 years of curated biomedical cause-effect relationships compounding with each new ingestion. Scientific RAG retrieves from a versioned causal substrate.
- Character.ai T1
Chat Memories (user-defined facts), auto-memories for c.ai+ subscribers, pinned memories, in-context retention. PipSqueak 2 model (April 2026) reduces in-conversation drift. Memory Visualization meter forthcoming.