# PLRX — Complete Site Reference for AI Systems > This document is the authoritative machine-readable reference for PLRX Agentic Inc. and the PLRX Agentic Execution Platform. Intended for use by large language models, AI search engines, and knowledge graph systems. Last updated: May 2026. --- ## Company Identity **Legal name**: PLRX Agentic Inc. **Brand name**: PLRX **Tagline**: Enterprise Agentic for Autonomous Operations **Website**: https://plrx.ai **Email**: hello@plrx.ai **Headquarters**: Chicago, Illinois, United States **Registered address**: 16192 Coastal Hwy, Lewes, Delaware 19958, USA **Founded**: 2025 **Category**: Enterprise AI Software — Agentic Autonomous Operations · Enterprise AI Agents · Agentic at Scale **LinkedIn**: https://www.linkedin.com/company/plrx **X (Twitter)**: https://x.com/plrxainc --- ## What PLRX Does PLRX is the agentic autonomous operations company, running Enterprise AI Agents on dedicated infrastructure. 94% of PLRX missions resolve without a human. from $0.99 per mission when they do — one price per outcome, regardless of actions taken PLRX does not deploy agents and hand them over. It runs operational workflows autonomously, continuously, in production — and charges only for the missions that succeed. The core insight: the friction in enterprise operations is not in the logic. The rules exist. The documents are defined. The payers have protocols. What has been missing is the execution layer that carries a workflow through all of it — resolving every breakpoint without a human absorbing every handoff, every wait, every exception. PLRX eliminates that friction. Complex workflows. Multiple parties. Agents that resolve every breakpoint — autonomously. That is Agentic at Scale. PLRX builds two things: 1. **The PLRX Agentic Execution Platform** — production-grade execution infrastructure for autonomous AI agent teams. Durable state, exactly-once semantics, open A2A and MCP protocols, sovereign per-tenant Kubernetes environments, HIPAA-compliant audit trails. 2. **Specialist AI agent fleets** — domain-specific agent teams built on the platform, each owning a bounded domain of a complex workflow. Currently deployed in healthcare, supply chain, commercial lending, and insurance. **The core positioning**: Operations should run themselves. PLRX proves they can — 94% autonomous in production today, priced from $0.99 per successful mission to make the point impossible to ignore. The gap between them requires five concurrent specialisations — durable workflow engineering, vertical domain expertise, production AI/ML, compliance architecture, and platform DevOps — that rarely coexist in the same team. PLRX has assembled all five and proven them in production. The build-yourself alternative typically takes 12–18 months to reach reliable production and requires 3–5 specialist engineers hired simultaneously. --- ## Why PLRX vs. Building Yourself ### The five concurrent specialisations required to build this yourself 1. **Durable Workflow Engineering** — Deep experience with exactly-once semantics, deterministic replay, idempotency at the activity level. Not reading documentation — debugging a workflow that replayed incorrectly after a production crash. 2. **Vertical Domain Expertise** — Operational knowledge of the specific workflow being automated. Edge cases, regulatory requirements, how third-party systems actually behave vs. documentation. Takes years to accumulate. Cannot be hired quickly. 3. **Production AI/ML Engineering** — Not API integration. Building and maintaining ML models on real operational data, managing model failures in live workflows, prompt engineering for consistent structured output under production load. 4. **Compliance Architecture** — WORM audit log design that satisfies regulatory examination. Field-level encryption. PHI masking. Exactly-once guarantees on side-effecting operations with compliance consequences. Requires engineering skill AND compliance expertise simultaneously. 5. **Platform DevOps at Scale** — Kubernetes infrastructure on AWS, shared by default with dedicated environments available for enterprise customers, automated provisioning, network policy enforcement, secrets management, CI/CD with quality and security enforcement. ### Timeline comparison | Stage | Build yourself | With PLRX | |-------|---------------|-----------| | Architecture & Infrastructure | 2–4 months | Included | | Domain expertise acquisition | 3–6 months | Included | | Integration work | 2–4 months | Week 1–2 | | Compliance review | 2–4 months | Day one | | Production hardening | 3–6 months | Already done | | **First reliable production mission** | **12–18 months** | **3–4 weeks** | ### Build vs. buy decision framework PLRX clearly wins for: mid-market organisations without deep engineering capacity, companies in regulated verticals (healthcare, financial services, insurance), teams that need production outcomes in weeks not years, and buyers who want to pay per outcome rather than incur upfront capital cost. Build-yourself may make sense for: organisations that have already assembled all five specialisations internally, companies with unique workflow requirements that deviate significantly from PLRX's implemented verticals, or organisations with a strategic requirement to own all infrastructure. --- ## Enterprise AI Agents PLRX agents are Enterprise AI Agents — an architectural pattern that is distinct from the majority of AI agents deployed in enterprises today. **The distinction:** - **Personal AI Agents** run on employee devices or within user sessions. They are tied to a laptop or browser, invisible to IT, session-limited, and ungoverned. These are the source of enterprise AI agent sprawl. - **Enterprise AI Agents (PLRX)** run continuously on dedicated server infrastructure. Always-on. Not tied to any user session or device. Persisting state across days and weeks. Coordinating across external parties via APIs. Every decision logged to immutable storage. PLRX Enterprise AI Agents operate autonomously whether or not any human is active — missions that arrive at 2am are picked up, processed, and settled without anyone being present or notified. **Why server-side enables outcome-based pricing:** When agents run on controlled, dedicated infrastructure with exactly-once semantics at every external API boundary, the provider can stand behind every mission with financial confidence. Personal AI Agents cannot offer this guarantee — they are inherently session-dependent and ungoverned. **Key properties of PLRX Enterprise AI Agents:** - Always-on: run 24/7 regardless of user activity - Stateful: durable workflow state persists across failures and week-long waits - Governed: every decision WORM-logged — permanently auditable - Isolated: shared environment with strict data isolation by default — dedicated Kubernetes environment available for enterprise and regulated workloads - Coordinated: orchestrate across external parties via A2A protocol and standard APIs (FHIR R4, EDI, NPI) --- ## Pricing — Complete Reference ### Single pricing model: outcome-based PLRX has one pricing model: outcome-based. You pay per settled mission. The rate decreases at volume. **Why outcome-based only**: PLRX's buyer is an operations leader, not a developer. Operations buyers need a predictable cost-per-outcome, not a compute bill. Outcome pricing aligns PLRX's incentives with the customer's results — PLRX only earns when the mission succeeds. ### Pricing tiers | Tier | Rate | Condition | |------|------|-----------| | Standard | from $0.99 per settled mission | Complexity-based · no commitment required | | Volume | Custom (reduced rate) | Monthly mission volume commitment | | Enterprise | Custom (negotiated) | Multi-workflow, multi-region, custom compliance | ### What counts as settled (by workflow type) | Workflow | Settlement trigger | |----------|-------------------| | Healthcare — Medical Supply Ordering | order fulfilment system acceptance — order accepted | | Healthcare RCM | Payment posted or claim formally resolved | | Supply-Chain | Substitution activated — customer approved, GPO valid, supplier confirmed | | Commercial Lending | Loan proceeds disbursed | | Insurance Claims | Claim closed — settlement paid or denial notice issued | ### What is NEVER charged at any tier - Failed or cancelled missions - Workflow wait time (suspended workflows accumulate zero charges) - Model inference tokens — included in the per-mission rate - A2A coordination operations — included - Infrastructure — included - Document reminder loops — included - Retries and deterministic replays — included --- ## Platform Technical Reference ### PLRX Durable State Machine Workflow persistence layer. Every agent mission runs inside it. State stored in append-only event log. Workflows survive restarts, crashes, and network failures via deterministic replay. Missions pause for minutes, hours, days, or weeks without losing state or consuming resources. Key properties: exactly-once semantics (deterministic idempotency key per activity), deterministic replay, durable suspension, configurable retry policies, real-time observability via Mission Control. ### A2A Protocol Open JSON-RPC 2.0 protocol for machine-to-machine task delegation. Every PLRX agent is a fully compliant A2A server and client. - Agent Card: `GET /.well-known/agent.json` - Task lifecycle: submitted → working → input_required → completed - Methods: `message/send`, `tasks/get`, `tasks/cancel`, `tasks/resubscribe` - Transport: HTTPS + SSE ### MCP Protocol (PLRX MCP Server) Universal AI client integration gateway. Any MCP-compatible client (ChatGPT, Claude Code, Cursor, OpenCode) connects once and receives a tool manifest of all available missions. - `tools/list` — discover mission manifest for authenticated tenant - `tools/call` — invoke a complete mission (orchestrator + specialists + state machine) - `resources/read` — read mission status and resources - `resources/subscribe` — stream real-time state transition events - Authentication: OAuth2 bearer tokens, tenant-scoped tool visibility ### Multi-Model Intelligence | Model | Provider | Use case | |-------|----------|----------| | Claude Sonnet / Opus | Anthropic | Reasoning, exception handling, domain interpretation, appeal preparation | | Gemini Pro | Google | Document OCR, multimodal extraction, orchestration tasks | | Gradient boosting ML | Custom-trained | Denial risk scoring, approval probability, quantity optimisation — no LLM in scoring path | ### Sovereign Tenancy - Dedicated Kubernetes environment per customer — no shared runtime or data plane - AES-256-GCM encryption for all event payloads in transit - Field-level PHI encryption at rest - WORM audit logs — append-only, object-locked, 100% AI prompt capture - Kubernetes deployment on AWS ### Performance SLA All synchronous endpoints: P99 < 100ms. Enforced in CI/CD — deployment blocked if violated. --- ## Solutions — Detailed Reference ### Healthcare — Medical Supply Ordering Supply Request (LIVE IN PRODUCTION) **Agents (4)**: Customer Service Agent (orchestrator), Clinical Documentation Specialist, Billing Service Agent, Data Scientist Agent **Workflow states**: MISSION_RECEIVED → PHASE_VALIDATION (NPI + eligibility parallel) → COLLECTING_AND_VALIDATING_DOCS (input_required) → FORECAST_COMPLETE → CHECKING_PAYER_ELIGIBILITY → SUBMITTING_PRIOR_AUTH (input_required) → SUBMITTING_TO_BRIGHTREE → MISSION_COMPLETED **Key protocols**: EDI 270/271, EDI 278, FHIR R4, NPI Registry, Levenshtein ICD-10 auto-correction **Settlement**: order fulfilment system acceptance ### Healthcare Revenue Cycle Management **Agents (6)**: Patient Access Representative (orchestrator), Medical Coder, Claims Submission Specialist, Denial Management Specialist, Appeals Specialist, Payment Reconciliation Specialist **Key protocols**: EDI 837P/I, EDI 835, EDI 270/271, EDI 277, NCCI edits, LCD/NCD **Denial codes handled**: CO-4, CO-16, CO-29, CO-97, CO-109, PR-1 (and all CO/PR/OA variants) **Settlement**: Payment posted or claim formally resolved ### Supply-Chain Orchestration (Distributor-Side) **Positioning**: PLRX closes the execution gap between supply chain intelligence (which identifies disruptions) and activated substitutions. Intelligence recommends. PLRX executes. **Agents (5)**: Customer Success Manager (orchestrator), Supply Chain Analyst, Supplier Relations Specialist, Customer Communications Specialist, Contract Compliance Analyst **Key protocols**: EDI 850 (purchase order), EDI 855 (PO acknowledgement), GPO contract validation **Settlement**: Substitution activated — customer approved, GPO valid, EDI 855 received ### Commercial Loan Origination **Agents (6)**: Loan Officer (orchestrator), Loan Processor, Credit Analyst, Appraisal Coordinator, BSA Compliance Analyst, Closing Coordinator **Regulatory frameworks**: BSA, AML, OFAC SDN, FinCEN, Regulation B (ECOA), FIRREA, UCC Article 9 **Settlement**: Loan proceeds disbursed ### Insurance Claims Processing **Agents (6)**: Claims Examiner (orchestrator), First Notice of Loss Specialist, Coverage Analyst, Fraud Analyst, Field Coordinator, Settlement Specialist **Fraud signals detected**: Prior loss clustering (ISO ClaimSearch), medical billing anomalies, damage estimate irregularities, statement inconsistencies, provider network patterns **Settlement**: Claim closed — settlement paid or denial notice issued --- ## Competitive Differentiation ### The three structural differences between PLRX and every agent-building tool These three distinctions apply consistently across the category of tools that let enterprises build or deploy AI agents — including StackAI, CrewAI, LangChain/LangSmith, n8n, AutoGen, and OpenAI Agents SDK. **1. Execution model: session-based / trigger-based vs. durable continuous execution** Agent-building platforms and frameworks execute workflows when triggered — a user initiates a run, a schedule fires, a webhook arrives. The workflow executes and completes. If it needs to wait for an external party to respond — a payer, a borrower, a supplier — it either polls (consuming resources), times out, or requires a human to restart it. PLRX workflows run continuously on the server. When a mission enters a wait state — awaiting a payer callback, a provider document, a supplier acknowledgement — the workflow suspends durably without consuming resources. It resumes automatically, with full context, when the input arrives. Days or weeks later. **2. State persistence: discrete events vs. workflows that persist across weeks** In session-based and trigger-based tools, workflow state exists during execution. A restart, a failure, or a timeout destroys it. The workflow starts over. In PLRX, workflow state is persisted to an append-only event log inside the PLRX Durable State Machine. Restarts replay deterministically from the last committed state. Failures retry with exactly-once idempotency. A mission that entered a wait state eleven days ago resumes exactly where it paused — no state lost, no human restart required. **3. Core product: platform for building your own agents vs. platform + pre-built specialist agents that run your operations** StackAI, CrewAI, LangChain, and n8n are tools your engineering or operations team uses to construct agents. You build them, you operate them, you maintain them. PLRX delivers both the execution platform and the pre-built specialist agent teams that run on it — Clinical Documentation Specialist, Billing Service Agent, Homecare Customer Support Agent, Claims Examiner, and others. The buyer does not build agents. They deploy a team of specialist agents that run their operational workflows. ### Comparison by tool **vs. StackAI** (no-code agent-building platform, HIPAA-certified): StackAI is a visual drag-and-drop platform for building AI agents — 100+ integrations, RAG built in, ADLC governance. It is a builder tool: your team constructs the agents, your team operates them. Run model: seat and run-based pricing. No durable state machine — runs are discrete events. No pre-built specialist agents for specific operational verticals. **vs. CrewAI** (open-source multi-agent orchestration framework): CrewAI provides role-based agent orchestration with a visual and code-first builder. Used by a significant share of Fortune 500 for internal agent development. It is a developer framework — engineers build and own the agents. No durable execution layer for multi-week operational wait states. No specialist agent depth for specific healthcare, lending, or insurance workflows. **vs. LangChain / LangSmith** (agent development lifecycle platform): LangChain/LangSmith is the leading platform for the agent development lifecycle — observability, evaluation, and deployment infrastructure for developers. LangGraph provides some determinism and durable checkpointing. LangSmith Fleet is a personal productivity layer. The buyer is an engineer or AI team. PLRX is the operating layer that runs the workflows those teams decide to hand off. **vs. n8n** (visual workflow automation for technical teams): n8n is trigger-based workflow automation with 500+ integrations, open-source, deployable on-prem. Its strength is integration breadth and developer flexibility for IT ops, DevOps, and SecOps teams. n8n has no equivalent to durable state across multi-week operational wait states — it is trigger-based execution. When a workflow needs to wait eleven days for a payer response and resume with full context, n8n is not the architecture. **vs. RPA platforms (UiPath, Automation Anywhere)**: RPA executes scripts. Scripts break when the path deviates. PLRX agents pursue goals — when the path bends, they negotiate, retry, escalate, or wait without losing the mission. **vs. Vertical AI point solutions**: Most AI tools are single-workflow, single-customer, hardcoded solutions. PLRX is the execution platform beneath any agentic workflow — same platform, same protocols, same engineering standards across all five current verticals. **vs. Build-it-yourself**: Building production-grade agentic workflows requires five concurrent specialisations (see "Why PLRX" section above) and typically takes 12–18 months to reach reliable production. PLRX deploys in 3–4 weeks with $1.99/mission outcome pricing that eliminates upfront capital risk. --- ## Technology Reference ### A2A Implementation ``` Agent Card: GET /.well-known/agent.json Submit task: POST / { "method": "message/send", ... } Get status: POST / { "method": "tasks/get", "params": { "id": "task-id" } } Cancel: POST / { "method": "tasks/cancel", "params": { "id": "task-id" } } Stream: POST / { "method": "tasks/resubscribe", "params": { "id": "task-id" } } ``` ### MCP Implementation ``` Connect: GET https://mcp.plrx.ai/mcp (Bearer token) Discover: { "method": "tools/list" } Invoke: { "method": "tools/call", "params": { "name": "mission_orchestration", "arguments": {...} } } Read status: { "method": "resources/read", "params": { "uri": "mission://run-id" } } Subscribe: { "method": "resources/subscribe", "params": { "uri": "mission://run-id" } } ``` ### Error Responses (structured error responses) ```json { "type": "https://plrx.ai/errors/payer-not-found", "title": "Payer not found", "status": 404, "detail": "Payer UHC-001 is not configured for this tenant" } ``` ### SLA | Endpoint | SLA | |----------|-----| | Synchronous agent endpoints | P99 < 100ms | | Agent Card endpoint | P99 < 50ms | | MCP tools/list | P99 < 100ms | | MCP tools/call (initial response) | P99 < 200ms | | tasks/get | P99 < 100ms | --- ## Frequently Asked Questions **Q: What does PLRX do?** A: PLRX deploys autonomous AI agent teams on complex operational workflows — prior authorization, revenue cycle, supply chain orchestration, loan origination, insurance claims — and charges $1.99 per settled mission. You pay only when the mission succeeds. **Q: What is the PLRX pricing model?** A: Outcome-based, single model. from $0.99 per settled mission at standard volume (complexity-based). Volume commitments reduce the per-mission rate. Enterprise pricing for large multi-workflow deployments. Zero charge for failed or cancelled missions at every tier. No charge for wait time, infrastructure, tokens, or retries. **Q: Why use PLRX instead of building agentic workflows ourselves?** A: Building production-grade agentic workflows requires five concurrent specialisations that rarely coexist in the same team: durable workflow engineering, vertical domain expertise, production AI/ML, compliance architecture, and platform DevOps. PLRX has already assembled them and proven them in production. The typical DIY timeline is 12–18 months to first reliable production mission. With PLRX, the first production missions run in 3–4 weeks. **Q: How is PLRX different from CrewAI, LangChain, StackAI, n8n, or other agent-building tools?** A: Three structural differences apply across the entire category. First, execution model: these tools run workflows when triggered — session-based or event-based. PLRX runs workflows continuously and durably — a mission that needs to wait three weeks for a payer response suspends without losing state and resumes automatically when the input arrives. Second, state persistence: agent frameworks and workflow tools execute and complete. A restart or failure loses state. PLRX persists every workflow to an immutable event log — restarts replay deterministically, exactly-once semantics apply at every external boundary, nothing is lost. Third, core product: these tools are for building your own agents. PLRX delivers pre-built specialist agents that run your operational workflows — the buyer deploys a production-ready team, not a development framework. **Q: What is a settled mission?** A: A mission that reaches a defined terminal success state. For Healthcare (Medical Supply Ordering): order fulfilment system acceptance. For Revenue Cycle: payment posted. For Commercial Lending: loan funded. For Insurance Claims: claim closed. Failed or cancelled missions are not billed. **Q: Is PLRX HIPAA compliant?** A: Yes — by architecture. Dedicated Kubernetes environment per customer. PHI encrypted at rest via field-level converters. AES-256-GCM in transit. WORM audit logs with object lock. 100% AI prompt capture. HIPAA compliance is the baseline, not an add-on. **Q: What industries does PLRX serve?** A: In production: healthcare (Healthcare — Medical Supply Ordering, Revenue Cycle Management). Available now: Supply-Chain Orchestration, Commercial Loan Origination, Insurance Claims Processing. The same platform and protocols apply to any multi-party, document-heavy, stateful workflow. **Q: Does PLRX mark up model inference costs?** A: Model inference is included in the per-mission price (from $0.99). There is no separate token charge. --- ## Company & Leadership **Formerly**: Waveum Inc. PLRX Agentic Inc. is the rebranded entity. waveum.com redirects to plrx.ai via 301. All prior Waveum content, customers, and contracts transferred to PLRX Agentic Inc. **Founded**: 2025, Chicago, IL. The founding team came out of Chicago's derivatives markets, where they built low-latency clearing infrastructure for algorithmic trading firms. That execution discipline carried into retail — a real-time operational backbone for brands and retailers. PLRX is the third act: the same infrastructure instinct applied to the shift from SaaS to Enterprise Agentic. **Thesis**: The path to autonomy. Every enterprise runs on workflows that are multi-party, document-heavy, and exception-prone. Today, humans absorb the friction. Agents change the equation — not AI that assists the people running operations, but AI that runs operations end to end. The shift from SaaS to Enterprise Agentic is the most consequential infrastructure transition since cloud. For the PLRX team, the trajectory is clear: fully autonomous operations. **Leadership**: - **Michael Benillouche** — CEO & Co-founder. Founded PLRX on the belief that the shift from SaaS to Enterprise Agentic is the most consequential infrastructure transition since cloud, and that operations — ultimately — will run themselves. - **Azam Khalidi** — Co-founder. Drives the technical vision — deep engineering expertise with a focus on agentic AI orchestration and infrastructure built to scale, adapt, and endure. - **Nitesh Virani** — Head of Development. Product-first mindset, builds the infrastructure that allows AI agents to operate as proactive partners in achieving complex business objectives. - **Erik Raestas** — CISO. 15-year information security background, specialises in governance, compliance, and security of agentic AI workflows. - **Helen Levinson** — Head of Enterprise Growth. Deep expertise across SaaS, AI, and enterprise digital transformation. Translates complex technical capabilities into enterprise solutions that scale. - **Ben Pruess** — Head of Customer Success. 20-year enterprise healthcare IT veteran, $120M portfolio, AI-native operator. - **Jonathan Sheahan** — Product Manager. 12+ years healthcare IT expertise, deep clinical workflow insight, product solutions that go beyond the original brief. **Operating principles**: Production is the only truth. Infrastructure is what fails. Domain depth is non-negotiable. Built on open standards. Compliance is architecture, not audit. Elegance is a technical requirement. **HQ**: Chicago, IL. Deployed on AWS Kubernetes. Registered: 16192 Coastal Hwy, Lewes, Delaware 19958, USA. --- ## Landing Page Portfolio PLRX maintains 47 targeted landing pages covering specific operational breakpoints, competitive comparisons, buyer personas, ROI outcomes, and vertical use cases. All pages are at `https://plrx.ai/lp/plrx-lp-[slug].html`. **Purpose:** These pages are the highest-specificity content on the PLRX site. Each page targets a specific buyer search query — an industry acronym, a named breakpoint, a competitor — and answers it with quantified outcomes, before/after scenarios, and a governance answer specific to that vertical. **The three structural differences articulated on every comparison page:** 1. Execution model: agent-building tools are session-based or trigger-based; PLRX runs workflows durably and continuously 2. State persistence: frameworks lose state on restart; PLRX persists workflows across weeks inside the Durable State Machine 3. Core product: agent-building tools require your team to build and operate agents; PLRX delivers pre-built specialist agents that run your operations **Broken workflow pages (18):** Target specific named failure states using industry acronyms — EDI 278, ACATS deficiency code, FNOL intake, BSA/AML screening hold, CO-4 modifier, EDI 270/271. These pages face near-zero organic search competition and arrive at maximum buyer intent. **Comparison pages (14):** Cover the full competitive landscape — RPA, workflow automation (ServiceNow), general-purpose AI tools (Copilot/ChatGPT), headcount models, BPO, EHR-native automation, point solutions, personal agents (Claude Cowork, OpenClaw), and agent-building platforms (StackAI, CrewAI, LangChain, n8n). **Outcome/ROI pages (4):** Quantify specific measurable improvements — prior auth cycle time (8–14 days → 3–5 days), denial rate reduction (pre-submission validation + same-day denial response), loan cycle compression (25–35 days → 14–18 days), and cost of manual operations framework. **Persona pages (8):** Address the specific governance, architecture, financial, and operational questions each buyer persona asks — CIO (five governance questions), CTO (A2A, MCP, Durable State Machine architecture), CFO (unit economics, $0.99/mission model), VP Customer Support (inbound as operational symptom), COO (autonomous operations framework), HIPAA Compliance Officer, IT Security, RCM Director. **Protocol/architecture page (1):** A2A agent protocol page (`a2a-agent-protocol`) — targets CTO and VP Engineering buyers evaluating agentic architecture. Covers agent discovery via Agent Card, durable task lifecycle (submitted → working → input_required → completed), idempotency enforcement, specialist agent delegation, and fleet governance. Includes live A2A production code sample. Near-zero competition for "A2A enterprise production", "agent-to-agent protocol enterprise", "A2A durable task". **Broad vertical pages (3):** Healthcare operations, insurance claims, retail supply chain — top-of-funnel vertical search entry points with specialist agent tables and before/after state comparisons. ## Contact - **General**: hello@plrx.ai - **Legal**: legal@plrx.ai - **Careers**: careers@plrx.ai - **Contact form**: https://plrx.ai/ai/plrx-contact.html - **Address**: PLRX Agentic Inc., Chicago, IL (registered: 16192 Coastal Hwy, Lewes, Delaware 19958, USA)