Platform Copilots Solutions Customers Resources Security Book a Demo →
YC Backed by Y Combinator · Built by Google AI leads

The operating system for value-based care.

Pelica unifies your claims, EHR, pharmacy, lab, and ADT data into one live record, and puts an AI copilot next to every team that depends on it. Risk adjustment, Stars, pharmacy, network, care management. Real-time.

One canonical record.
Every workflow. Real-time.

Three layers. Each addressable on its own. Deployed together, the equivalent of 8 to 15 point vendors in a single platform.

01
Live data
Claims (MMR, MAO-004, MOR, 277CA), EHR (FHIR/HL7), pharmacy (PDE), labs, ADT, telephony recordings, payer SFTP drops. Built for ECDS volumes: 35-75x the supplemental data of hybrid sampling. Tenant-isolated, BAA-covered, encrypted in transit and at rest.
02
The AI workforce
A copilot per role plus a cross-cutting AI Data Analyst. Every recommendation cites its evidence. Trumping logic, RADV-defensibility, and submission-window awareness built in.
03
Action
Outbound voice (TCPA-compliant, multilingual). EMR overlays (Epic SMART-on-FHIR, Athena marketplace). Provider and staff portal. Coder workspace with encounter-and-claim review.

A copilot for every team.
One canonical record they all read from.

Each copilot is independently useful. Because they read and write to the same record, an HCC captured by Risk informs the Network agenda; a Part D gap closed by Pharmacy updates the Stars projection.

Risk Adjustment

Real-time HCC capture, RADV-defensible.

Quality & Stars

Unified gap closure across every payer.

Pharmacy & Part D

PDC lift on the three triple-weighted measures.

Provider Network

EMR overlay. Providers get credit for care already delivered.

Care Management

ADT-driven TOC worklists. No duplicate calls across teams.

AI Data Analyst

Featured below

Plain-English answers from the canonical record.

app.pelica.ai / analyst
New analysis
Stars glide-path · Q2
PDC cohort review
Live · canonical record
Ask Pelica · en · plain English to SQL
Ran in 1.8s V28-aware Trumping logic View SQL →
Members
1,284
Aetna · MA panel
Open HCCs
2,471
+18 today
Projected lift
$3.1M
PMPY · current contract
Window closes
42d
MAO-004 · Q2 cycle
Top opportunities 7 of 1,284 · scroll for more
Export Route to coder queue
Member HCC Action Lift
MR
M. Ruiz
MBI-•••4821
HCC 37 · DM w/ comp PCP recapture $1,847
DA
D. Adler
MBI-•••0193
HCC 36 · DM uncontrolled Care mgmt TOC $1,212
TT
T. Truong
MBI-•••7340
HCC 18 · CHF Specialist note $1,054
RV
R. Vasquez
MBI-•••2718
HCC 111 · COPD PCP recapture $908
SO
S. Okeke
MBI-•••5562
HCC 85 · CHF chronic Encounter review $842
JN
J. Nguyen
MBI-•••9104
HCC 38 · DM w/o comp Outreach · A1C $711
AP
A. Patel
MBI-•••3027
HCC 22 · Morbid obesity Documentation fix $612

Three pains. One root cause.
The same OS.

Quality, risk, and care leaders across MA plans, ACOs, IPAs, and medical groups describe the same three pains. They're structural, not vendor-specific, and each one gets fixed at a different layer of the work.

"Our data lives in twelve places. By the time we know there's a problem, the submission window has closed."
DR Director of Risk Adjustment
The fix lives in the data layer
"One team contacts a member for a care gap another team already closed. We're solving the same problem five times."
SM Stars Program Manager
The fix is one shared record
"We can build the most beautiful predictive model. If we can't get it into the provider's hands, it's nothing."
VQ VP of Quality
The fix lives in the action layer

How a physician-led IPA stopped
calling the same patient three times.

A physician-led IPA managing 175,000 commercial, Medicare, and Medicaid members across New York. Here's what changed when their teams started working from one shared record.

Physician-led IPA · New York

One canonical record across risk, quality, pharmacy, and care management.

A physician-led IPA serving 175,000 commercial, Medicare, and Medicaid members across New York. Risk adjustment ran out of one set of files. Quality and Stars ran out of another. Pharmacy adherence lived in the PBM portal. Provider performance was assembled in spreadsheets.

Patients fell through the seams between teams. The same member got three calls in the same week: one from risk, one from quality, one from the practice. Sometimes in a language the member didn't speak. Submission windows closed before suspect HCCs surfaced.

The mandate

One shared record across risk, quality, pharmacy, and care management. Outreach in the right language, only once. RADV-defensible chain-of-custody on every captured HCC. No new headcount.

In their words
"We were calling the same member three times. Risk, then quality, then the practice. Sometimes in a language they didn't even speak. Now one outreach goes out, in the right language, and the patient actually comes in."
Quality Coordinator New York

For the teams that own
risk-bearing contracts.

If your margin moves with clinical and quality outcomes, your team is the buyer. We work across health plans, IPAs and MSOs, ACOs, and risk-bearing medical groups.

Health plans

MA · Dual-SNP · Regional Blues

COO · SVP Stars · SVP Risk Adjustment · SVP Pharmacy

IPAs and MSOs

Delegated risk-bearing

COO · Director of Risk Adjustment · Director of Quality · VP Provider Network

ACOs

MSSP · REACH · CINs

ACO Executive Director · VP Population Health · CMO

Medical groups

Primary care · Multi-specialty · DSOs

CMO · COO · VP Operations

Built by AI leaders from Google.
Focused on patients.

Two ex-Google AI engineers, embedded directly with VBC operators.

Catherine Zhao, Co-Founder and CEO of Pelica Health

Catherine Zhao

Co-Founder & CEO

Led Google AI work on Gemini and enterprise AI agents. Clinical AI researcher at Stanford School of Medicine. Front-line patient care at clinics in California and New Hampshire.

Google AI · Gemini Stanford Research Dartmouth Y Combinator
Lalit Kundu, Co-Founder and CTO of Pelica Health

Lalit Kundu

Co-Founder & CTO

Built software for regulated industries at Google and YouTube. Engineering and ML lead on YouTube Payments, running a 45+ person team in Commerce Billing. Wharton MBA. ACM-ICPC World Finalist.

Google · YouTube ML Wharton MBA ACM-ICPC Finalist Y Combinator

Backed by Y Combinator and operators from healthcare and AI companies.

Y Combinator SurgePoint Capital Upside Partnership ISEED Ventures Progressive Ventures

Built for healthcare's highest compliance bar.

Tenant isolation, encryption end-to-end, full audit trail, and contractual zero-retention with every LLM provider in use. Available on request: SOC 2 Type II report, BAA template, subprocessor list, pen-test history.

Visit our Trust Center
AICPA SOC 2 Type II compliance badge
SOC 2 Type II
Compliant
HIPAA compliance badge
HIPAA
Compliant
SOC 2 Type II Compliant
Independently audited on an annual cadence. Customer data is protected under continuous operational and security controls.
HIPAA Compliant. BAA available.
Business Associate Agreement signed at deployment. PHI handled under full HIPAA Security and Privacy Rule compliance.
Tenant isolation. End-to-end encryption.
Per-customer data plane. Encryption at rest and in transit. Customer-managed keys, role-based access, automated backups, and zero-retention agreements with every LLM provider in use.
Full audit trail. SAML / SSO.
Every recommendation, query, and outbound action logged with attribution. SAML / SSO via Identity Platform. HITRUST certification in progress.
See it in action

See Pelica on data that looks like yours.

30-minute walkthrough. We'll run the copilots live on a redacted book of business from a real risk-bearing org.

Book a Demo →