Terms of Service

1. Acceptance of Terms & Partnership Foundation

By commissioning, accessing, or integrating any automated systems, agent instances, MCP connectors, or workflow pipelines engineered by Operon Core Systems Inc. (“Operon Core”, “we”, “us”), you (“Client”, “you”) agree to be legally bound by these Terms of Service. Operon Core operates not as a detached vendor, but as a dedicated, long-term AI engineering partner founded on mutual trust, transparency, and mission alignment.

2. Scope of Services & Sprint Deployments

Operon Core develops, delivers, and maintains bespoke autonomous AI agent networks governed by the Operon Core™ Engine (including Orin, Nova, Echo, Vox, and custom multi-agent orchestrators). Standard builds follow structured sprint delivery cadences (typically 48 to 72 hours for core agents and 5 to 14 business days for enterprise multi-agent networks), documented with milestones and sandbox test environments.

3. Payment Structure & Milestone Billing (50% Upfront / 50% Delivery)

To guarantee senior engineering resource allocation and dedicated infrastructure provisioning from Day 1, all project implementations adhere to the following milestone billing model:

• 50% Upfront Commitment Deposit: Invoiced and due upon agreement execution prior to project kickoff, architectural modeling, and sprint initiation.
• 50% Completion Balance: Invoiced upon milestone delivery, end-to-end sandbox verification, and formal client sign-off, prior to full production DNS/telephony cutover.

4. Monthly Retainers & Rapid After-Service Commitment

Deploying an autonomous agent is merely the first milestone. The true return on investment depends on continuous, proactive after-service and operational stewardship, which is a core pillar of Operon Core:

• Rapid Response Priority SLA: Active monthly retainer clients receive priority support with sub-60 minute engineering triage for critical pipeline blocks during standard operational windows.
• Proactive Telemetry & Latency Watch: 24/7 automated synthetic pinging and observability to detect and mitigate third-party API latency spikes or deprecations before they impact end customers.
• Prompt Hardening & Model Upgrades: Continuous tuning, few-shot prompt adjustments, vector index re-indexing, and seamless migration to newer foundation model checkpoints without downtime.
• Dedicated Engineering Channel: Direct communication through private Slack or WhatsApp channels for rapid iterations and ad-hoc operational tweaks.

5. Strict No-Refund Policy

Due to the immediate allocation of dedicated senior AI engineers, non-recoverable cloud compute and GPU provisioning, third-party API provisioning, and bespoke intellectual architecture created specifically for Client from the first hour of engagement, all upfront deposits, milestone payments, and retainer fees are strictly non-refundable once project work or a billing cycle commences. We encourage complete review during the initial audit call to ensure full alignment before commissioning a build.

6. Enterprise Security, Data Isolation & Zero LLM Training

Data security and privacy are non-negotiable foundations of Operon Core's architecture:

• Zero Foundation Model Training: Under no circumstances is Client data, proprietary knowledge base content, customer conversation logs, or internal documents used to train, fine-tune, or improve public foundation models (OpenAI, Anthropic, Google, etc.).
• Encryption & Secret Isolation: All sensitive credentials (API tokens, database strings, CRM keys) are vaulted with AES-256 encryption at rest and isolated within private client environments. All network transit is enforced over TLS 1.3.
• Tenant Sandboxing: Each client's agent runtime, vector stores, and automated execution pipelines are strictly isolated to prevent cross-tenant data leakage.

7. Agentic AI Operational Rules & Guardrails

To guarantee safe, deterministic, and legally compliant autonomous systems:

• Deterministic Guardrails & Fallbacks: Agents operate within defined JSON schemas, deterministic tool definitions, and strict behavioral bounds to eliminate hallucinations.
• Human-in-the-Loop Escalation: Every agent is equipped with automated escalation triggers. Whenever intent ambiguity, sentiment volatility, or high-consequence thresholds are detected, conversations transfer smoothly to human personnel.
• Acceptable Use & Compliance: Client warrants that automated agents will not be utilized for deceptive practices, unsolicited spamming, predatory outreach, harassment, or activities violating telecommunication guidelines (including Twilio A2P 10DLC, Meta WhatsApp Business policies, and CAN-SPAM regulations).

8. Client Ownership & Intellectual Property

Upon receipt of full payment for each sprint or milestone, Client retains 100% intellectual property ownership of all custom prompts, business workflow recipes, n8n/Make scenarios, database integration scripts, and domain-specific configuration created for Client. Operon Core retains ownership of its underlying platform engine, reusable framework primitives, and proprietary scaffolding libraries.

9. Trust, Integrity & Relationship Commitment

Long-term business value in agentic AI is achieved through authentic relationships and iterative excellence. Operon Core commits to transparent reporting, realistic engineering estimates, clear system telemetry, and proactive problem-solving. In return, Client commits to prompt feedback, timely milestone sign-offs, and open collaboration.

10. Governing Law & Inquiries

These terms are governed by the laws applicable to commercial software agreements. Questions regarding legal terms, custom enterprise MSAs, or data protection agreements may be directed to our executive team at [email protected].