# Runtime > Runtime is the AI agent harness for payment and fintech teams. It is multi-cloud, multi-sandbox, and multi-model, so agents never depend on one vendor. From payment ops to risk, compliance, and support, anyone on your team can build and run AI agents that work like teammates. Each agent runs on its own computer in your cloud, is called from Slack, Teams, email, SMS, or voice, and works with role-based access, human approvals, and a full audit trail. Runtime is also an applied AI lab for payment and risk operations (https://runtm.com/lab), with forward-deployed AI engineers who build the first agents with each customer. ## What is Runtime? Runtime puts your procedures into action with specialized agents for payment and risk ops. Give each agent a defined job, approved tools and data, and clear rules for actions, approvals, and escalations. Agents are triggered from Slack, email, GitHub, tickets, or a schedule. They investigate, do the work, and report back, with sensitive actions held for a human sign-off. Every run is stored with its steps, tools, cost, and result, so the company gets one system of record for the work its agents do. Teams can start with one workflow, such as payment investigations, partner-bank requests, merchant reviews, or reconciliation exceptions, and expand to other jobs on the same platform. Runtime covers four groups of work: getting customers live (underwriting, KYB/KYC, sales engineering, implementation), keeping money moving (payment ops, partner-bank requests, disputes, reconciliation), protecting the business (fraud, transaction monitoring, compliance), and supporting customers (payments support, escalations). ## How Runtime works 1. Create a specialized agent: define its job, procedures, and approval or escalation rules 2. Connect its tools and data: scope access to what the agent can read and do 3. Put it to work: trigger it from a request, alert, or schedule 4. Review and approve: inspect the evidence, activity, and outcome, and approve sensitive actions ## Bring your own tools and data Anything with an API, a CLI, or an MCP server becomes a tool the agent can use. Warehouses, databases, and internal services connect as data sources, read-only where it matters. Permissions are set once by engineering and enforced on every run. Common integrations include Zendesk, Intercom, Slack, Stripe, GitHub, Datadog, Sentry, PagerDuty, Snowflake, BigQuery, Postgres, HubSpot, Salesforce, Linear, and Notion. ## Works with the coding agents you already use Runtime hosts the environments, templates, and policies, and runs the agent your team already prefers: - Claude Code (Anthropic) - Codex (OpenAI) - Cursor CLI - Gemini CLI ## A system of record for every agent run - Every run is stored with the prompt, the steps, the tools used, the cost, and the result - Agents read their previous runs and get better at the job over time - One view of cost and activity across every agent, user, and team ## Security and control - Agents work against isolated environments with the scopes you assign, including PII redaction and row-level limits - Production writes happen only through reviewed actions or pull requests - Approval gates on refunds, writes, and anything else you mark sensitive - SOC 2 Type II audited, encrypted at rest and in transit - SSO and role-based access for the whole team - We never train on your data - Self-hosting available in your own AWS, GCP, or Azure account ## Who uses Runtime Runtime is built for payment and risk ops teams at payment companies on both sides of the transaction: issuers and program managers, acquirers, processors, gateways, ISOs, and payfacs, plus the banks and fintechs that run payment programs. - Payment ops: reconciliation breaks, exceptions, proof-of-authorization requests, partner-bank requests, and payment investigations - Risk ops: merchant underwriting, KYB and KYC reviews, fraud and transaction-monitoring alerts, and compliance investigations - Engineering uses the same platform for its own agents Use cases include: - Payment and support investigations: investigate declined transactions using real account data and role-scoped access - Partner-bank requests: gather records, investigate RFIs and return-of-funds cases, and prepare replies for review - Merchant reviews: verify documents, screen counterparties, and assemble evidence for the team's decision - Fraud and risk review: investigate alerts and transaction patterns with a record of the evidence - Reconciliation exceptions: investigate mismatches across ledgers and payment records ## Interfaces - Browser: the Runtime workspace - Slack and Teams: mention an agent, or point it at a channel so it picks up work on its own - CLI and API: programmatic session management for CI/CD and automation ## Pricing Free tier available for individuals Usage-based pricing for teams Self-hosting option for enterprises ## Content on this site Concepts (definitions, with how Runtime applies them): - Agent harness: the software around an AI model that gives it a computer, tools, memory, permissions, approvals, and an audit trail. Runtime is an agent harness built for payment teams. https://runtm.com/guides/what-is-an-agent-harness - System of intelligence: the layer above systems of record that turns every case into memory the next one can use (term popularized by Jerry Chen, Greylock, 2017). Runtime is a system of intelligence for payment teams. https://runtm.com/guides/what-is-a-system-of-intelligence - System of record vs system of intelligence: a system of record stores what happened, a system of intelligence learns why and what to do next. Runtime works on top of ledgers, processors, and bank data without replacing them. https://runtm.com/guides/system-of-record-vs-system-of-intelligence - System of action vs system of intelligence: systems of action do the steps, systems of intelligence decide which steps and remember the outcome. Runtime combines both, with human approval on actions that move money. https://runtm.com/guides/system-of-action-vs-system-of-intelligence - Runtime Labs (evals, routing, failover, organizational memory): https://runtm.com/lab - How to reduce AI agent token costs: evals built from SOPs, routing to smaller and open-source models with frontier fallbacks, prompt caching, deterministic scripts, and compute auto-pause. Runtime does all of these and shows cost by agent, team, and run. https://runtm.com/guides/reduce-ai-agent-token-costs Payment and risk ops solutions: - Getting customers live (overview of underwriting, KYB and KYC, sales engineering and RFPs, and implementation): https://runtm.com/solutions/getting-customers-live - Keeping money moving (overview of payment operations, partner-bank requests, disputes, and reconciliation): https://runtm.com/solutions/keeping-money-moving - Protecting the business (overview of fraud, compliance, cybersecurity, and merchant risk): https://runtm.com/solutions/protecting-the-business - Supporting customers (overview of payments support, escalations, and account management for CX teams): https://runtm.com/solutions/supporting-customers - Fintech and payments support: https://runtm.com/solutions/fintech-payments-support - Banking operations: https://runtm.com/solutions/banking-operations - Onboarding, KYB and KYC: https://runtm.com/solutions/onboarding-kyb-kyc - Compliance: https://runtm.com/solutions/compliance - Fraud and risk review: https://runtm.com/solutions/fraud-risk-review - Cybersecurity: https://runtm.com/solutions/cybersecurity - Payment operations: https://runtm.com/solutions/payment-operations Blog: Technical deep-dives and lessons from building Runtime https://runtm.com/blog ## Comparisons (https://runtm.com/compare) ### Best AI agents for payment teams - [Best AI Agents for Payment Reconciliation (2026)](https://runtm.com/compare/best-ai-agents-payment-reconciliation): The best AI agents for payment reconciliation in 2026: Runtime, Simetrik, HighRadius, BlackLine, Trintech, Numeric, Ledge, and Duco compared on matching, exceptions, posting, and audit. - [Best AI Agents for Chargeback Management (2026)](https://runtm.com/compare/best-ai-agents-chargebacks): The best AI agents for chargeback management in 2026: Runtime, Chargeflow, Justt, Disputifier, Stripe, Chargebacks911, Verifi, Ethoca, and Midigator compared on evidence, alerts, and pricing. - [Best AI Agents for AML Alert Triage (2026)](https://runtm.com/compare/best-ai-agents-aml-alert-triage): The best AI agents for AML alert triage in 2026: Runtime, Unit21, Hummingbird, Flagright, Sardine, Hawk, Lucinity, Bretton AI, and WorkFusion compared on coverage, SARs, and audit. - [Best AI Agents for Merchant Underwriting (2026)](https://runtm.com/compare/best-ai-agents-merchant-underwriting): The best AI agents for merchant underwriting in 2026: Runtime, Ballerine, Middesk, Sardine, G2RS, Enigma, Persona, Footprint, and LexisNexis compared on KYB, website risk, and monitoring. ### Alternatives - [Simetrik Alternatives for Payment Reconciliation (2026)](https://runtm.com/compare/simetrik-alternatives): The best Simetrik alternatives for payment reconciliation in 2026: Runtime, Modern Treasury, Ledge, Optimus, Duco, and BlackLine, compared on fit, where they run, AI agents, and pricing. - [Sardine Alternatives for Fraud and Compliance (2026)](https://runtm.com/compare/sardine-alternatives): The best Sardine alternatives for fraud and compliance in 2026: Runtime, Unit21, Alloy, SEON, Feedzai, and Sift, compared on fit, AI agents, where they run, and published pricing. - [Sift Alternatives for Fraud Prevention (2026)](https://runtm.com/compare/sift-alternatives): The best Sift alternatives for fraud prevention in 2026: Runtime, Forter, Signifyd, Sardine, SEON, and Feedzai, compared on fit, guarantees, AI agents, where they run, and pricing. - [Footprint Alternatives for KYC and Risk Ops (2026)](https://runtm.com/compare/footprint-alternatives): The best Footprint alternatives for KYC and risk operations in 2026: Runtime, Alloy, Socure, Sumsub, Middesk, and Unit21, compared on fit, AI agents, where they run, and published pricing. ### Runtime vs alternatives - [Runtime vs Building Your Own Agent Platform](https://runtm.com/compare/build-in-house): Should a payment or fintech company build its own AI agent platform or buy one? What it takes to run mission-critical agents in-house, how long it takes, and where Runtime fits. - [Runtime vs Gumloop](https://runtm.com/compare/gumloop): Runtime is the agent harness for payment teams: agents investigate cases in your cloud, with approvals and an audit trail. Gumloop is a visual AI workflow builder for general business automation. - [Runtime vs Workato](https://runtm.com/compare/workato): Runtime is the agent harness for payment teams, built to investigate exceptions in your own cloud. Workato is an enterprise iPaaS adding AI agents on top of its integrations. - [Runtime vs Bretton AI](https://runtm.com/compare/bretton): Runtime is one agent harness for every payment team, from compliance to payment ops, finance, and support. Bretton AI builds AI agents and managed services for financial crime compliance. - [Runtime vs Footprint](https://runtm.com/compare/footprint): Runtime is one agent harness for every payment team, from risk to payment ops, finance, and support. Footprint is an identity and risk operations platform with AI agents for KYC, KYB, and fraud. - [Runtime vs Lindy](https://runtm.com/compare/lindy): Runtime is the agent harness for payment teams, with agents that take on sensitive workflows in your cloud. Lindy is a no-code platform for building AI assistants and agents. Compare fit. - [Runtime vs n8n](https://runtm.com/compare/n8n): Runtime is the agent harness for payment teams, with approvals, audit, and payments expertise built in. n8n is a source-available workflow automation tool that teams can self-host. - [Runtime vs UiPath](https://runtm.com/compare/uipath): Runtime is the agent harness for payment teams, with agents that reason through exceptions instead of following scripts. UiPath is an enterprise RPA and agentic automation platform. - [Runtime vs Zapier](https://runtm.com/compare/zapier): Runtime is the agent harness for payment teams, with agents that investigate exceptions in your cloud. Zapier is the most widely used no-code automation platform, now with AI agents. - [Runtime vs Decagon](https://runtm.com/compare/decagon): Runtime is one agent harness for every payment team, including support. Decagon builds AI agents for customer support across chat, email, and voice. Compare scope, data, and fit. - [Runtime vs HappyRobot](https://runtm.com/compare/happyrobot): Runtime is the agent harness for payment teams, with voice as one of several channels. HappyRobot builds voice-first AI agents, best known in logistics and freight. Compare fit. - [Runtime vs Sierra](https://runtm.com/compare/sierra): Runtime is one agent harness for every payment team, including support. Sierra builds conversational AI agents for customer experience. Compare scope, deployment, models, and fit. - [Runtime vs Rogo](https://runtm.com/compare/rogo): Runtime is the agent harness for payment and fintech operations teams. Rogo is an AI research platform for bankers and investors. Compare users, data, actions, and deployment. - [Runtime vs Runbear](https://runtm.com/compare/runbear): Runtime is the agent harness for payment teams, with agents that investigate on their own computers in your cloud. Runbear is a no-code way to put AI agents in Slack and Teams. - [Runtime vs Factory](https://runtm.com/compare/factory): Runtime is the agent harness for payment teams beyond engineering: ops, risk, compliance, and support. Factory builds Droids, AI agents for enterprise software engineering. Compare fit. - [Runtime vs Devin](https://runtm.com/compare/devin): Runtime is the agent harness for payment teams, with agents for operations work in your cloud. Devin, from Cognition, is an autonomous AI software engineer. Compare users, models, and fit. - [Runtime vs Lorikeet](https://runtm.com/compare/lorikeet): Runtime vs Lorikeet compared for fintech teams: Lorikeet's AI support concierge for regulated industries vs Runtime's agent harness for support, payment ops, finance, and risk in your cloud. - [Runtime vs Parahelp](https://runtm.com/compare/parahelp): Runtime vs Parahelp compared: Parahelp's fast-setup AI support agent for software companies vs Runtime's agent harness for fintech support, payment ops, finance, and risk in your own cloud. - [Runtime vs Forethought](https://runtm.com/compare/forethought): Runtime vs Forethought compared: Forethought's multi-agent AI for customer support, now part of Zendesk, vs Runtime's agent harness for fintech support, payment ops, finance, and risk. - [Runtime vs Sardine](https://runtm.com/compare/sardine): Runtime vs Sardine compared: Sardine is an agentic fraud and AML platform with ten risk agents. Runtime is an agent harness for every payment team, in your cloud. Where each fits, and how to run both. - [Runtime vs Simetrik](https://runtm.com/compare/simetrik): Runtime vs Simetrik compared: Simetrik is a reconciliation and financial control platform with the Simetrik Agent. Runtime runs agents for every payment team, in your cloud. Where each fits. - [Runtime vs Google Gemini Enterprise Agent Platform](https://runtm.com/compare/google-gemini-enterprise-agent-platform): Compare Runtime and Google Gemini Enterprise Agent Platform (formerly Vertex AI): agent building blocks on Google Cloud vs a finished agent harness for payment ops, risk, and finance teams. - [Runtime vs AWS Bedrock AgentCore](https://runtm.com/compare/aws-bedrock-agentcore): Compare Runtime and Amazon Bedrock AgentCore: modular AWS services for building and running agents vs a finished agent harness for payment ops, risk, and finance teams, in any cloud. - [Runtime vs LangChain](https://runtm.com/compare/langchain): Compare Runtime and LangChain: open-source agent frameworks (LangChain, LangGraph, Deep Agents) plus LangSmith Fleet and Deployment vs a finished agent harness for payment ops, risk, and finance. - [Runtime vs Notion Agents](https://runtm.com/compare/notion-agents): Runtime vs Notion Custom Agents for fintech teams: workspace knowledge agents vs agents that investigate payments in your cloud, with approvals on money movement and a sponsor-bank audit trail. - [Runtime vs Dust](https://runtm.com/compare/dust): Runtime vs Dust for fintech and payment teams: a horizontal platform for company-wide knowledge agents vs an agent harness for payment operations in your cloud, with approvals on money movement. - [Runtime vs Grok Bot](https://runtm.com/compare/grok-bot): Grok Bot gives you always-on AI teammates on a shared cloud computer for inbox, research, and apps. Runtime runs governed agents on payment data in your cloud. Compare scope, data, and pricing. - [Runtime vs Town](https://runtm.com/compare/town): Town gives each person a Townie, an AI assistant with its own email that learns how you work. Runtime runs governed agents on payment data for whole teams, in your cloud. Compare data and pricing. - [Runtime vs Hyperagent](https://runtm.com/compare/hyperagent): Hyperagent is a horizontal platform for teams of AI agents with their own computers, memory, and skills. Runtime is the agent harness for payment teams, in your cloud. Compare data, models, pricing. - [Runtime vs Convey](https://runtm.com/compare/convey): Convey builds digital teammates that operators train by sharing their screen, for finance, accounting, and ad ops. Runtime is the agent harness for payment teams, in your cloud. Compare the two. - [Runtime vs OpenAI Dots](https://runtm.com/compare/openai-dots): Runtime vs OpenAI Dots compared: an always-on personal agent in ChatGPT on GPT-6 Astra vs a multi-cloud, multi-model agent harness for payment ops, risk, finance, and compliance teams. - [Runtime vs Claude Managed Agents](https://runtm.com/compare/claude-managed-agents): Runtime vs Claude Managed Agents compared: Anthropic's hosted agent harness for Claude vs a multi-cloud, multi-sandbox, multi-model agent harness with guardrails and teams for payment operations. - [Runtime vs Instinct](https://runtm.com/compare/instinct): Instinct is an invite-only personal AI agent you text or call to run errands. Runtime runs governed agents for payment teams in your cloud, on any model, with guardrails you set. - [Runtime vs Meta Muse](https://runtm.com/compare/meta-muse): Meta Muse is a consumer AI agent that sends email, books travel, fills forms, and pays for things. Runtime runs governed agents for payment teams in your cloud, on any model, with guardrails you set. ### Head-to-head comparisons - [Lorikeet vs Decagon](https://runtm.com/compare/lorikeet-vs-decagon): Lorikeet vs Decagon compared for fintech and regulated support teams: channels, compliance, deployment, implementation, and pricing, plus where Runtime fits behind the ticket. - [Parahelp vs Decagon](https://runtm.com/compare/parahelp-vs-decagon): Parahelp vs Decagon compared: setup speed, channels, configuration, compliance, and pricing for AI customer support agents, plus where Runtime fits for fintech teams behind the ticket. - [Forethought vs Sierra](https://runtm.com/compare/forethought-vs-sierra): Forethought vs Sierra compared: support AI suite now part of Zendesk vs Sierra's branded agent platform. Channels, compliance, financial services fit, pricing, and where Runtime fits. - [Sardine vs Sift](https://runtm.com/compare/sardine-vs-sift): Sardine vs Sift compared for fintech and payments: fraud scoring, AML and KYC coverage, AI agents, network data, and pricing. Plus where Runtime fits as the agent layer for every payment team. - [Alloy vs Sumsub](https://runtm.com/compare/alloy-vs-sumsub): Alloy vs Sumsub compared for 2026: data orchestration vs all-in-one verification, KYC and KYB coverage, AI agents, case management, and pricing, plus where Runtime fits next to either. - [Middesk vs Sumsub](https://runtm.com/compare/middesk-vs-sumsub): Middesk vs Sumsub for KYB in 2026: US business identity depth vs global coverage, UBO checks, AI agents, monitoring, and pricing, plus where Runtime fits for merchant reviews. - [AWS Bedrock AgentCore vs Google Gemini Enterprise Agent Platform](https://runtm.com/compare/bedrock-agentcore-vs-google-agent-platform): AWS Bedrock AgentCore vs Google Gemini Enterprise Agent Platform compared for 2026: runtimes, memory, identity, policy, models, evaluation, and pricing, plus where Runtime fits for payment teams. - [LangChain vs AWS Bedrock AgentCore](https://runtm.com/compare/langchain-vs-bedrock-agentcore): LangChain (LangGraph, Deep Agents, LangSmith) vs AWS Bedrock AgentCore compared for 2026: frameworks vs managed services, deployment, observability, guardrails, and pricing, plus where Runtime fits. - [Salesforce Agentforce vs Sierra](https://runtm.com/compare/agentforce-vs-sierra): Salesforce Agentforce vs Sierra for financial services and fintech: product scope, CRM fit, channels, compliance, pricing, and where Runtime fits for payment operations behind the conversation. - [Dust vs Notion Agents](https://runtm.com/compare/dust-vs-notion-agents): Dust vs Notion Custom Agents: how the two workspace AI agent platforms compare on scope, connectors, models, governance, and pricing, and where Runtime fits for fintech payment operations. - [Grok Bot vs Town](https://runtm.com/compare/grok-bot-vs-town): Grok Bot vs Town: xAI's always-on AI teammates with their own cloud computer, compared with Town's personal assistant that learns how you work. Features, security, pricing, and where Runtime fits. - [Hyperagent vs Convey](https://runtm.com/compare/hyperagent-vs-convey): Hyperagent vs Convey: a self-serve platform for teams of AI agents compared with digital teammates trained by screen share. Features, security, deployment, pricing, and where Runtime fits for fintech. - [OpenAI Dots vs Claude Managed Agents](https://runtm.com/compare/openai-dots-vs-claude-managed-agents): OpenAI Dots vs Claude Managed Agents compared: an always-on personal agent in ChatGPT vs Anthropic's hosted agent harness for developers. Audience, sandboxes, controls, pricing, where Runtime fits. - [Claude Managed Agents vs AWS Bedrock AgentCore](https://runtm.com/compare/claude-managed-agents-vs-bedrock-agentcore): Claude Managed Agents vs Amazon Bedrock AgentCore compared for 2026: managed harnesses, sandboxes, models, permissions, memory, and pricing, plus where Runtime fits for payment and fintech teams. - [Instinct vs Meta Muse](https://runtm.com/compare/instinct-vs-meta-muse): Instinct vs Meta Muse: the invite-only AI agent you text or call, compared with Meta's personal agent in an app, on the web, and in WhatsApp. Capabilities, privacy, pricing, and where Runtime fits. - [Instinct vs OpenAI Dots](https://runtm.com/compare/instinct-vs-openai-dots): Instinct vs OpenAI Dots: the invite-only AI agent you text or call, compared with ChatGPT's always-on agents with their own cloud computer. Capabilities, controls, pricing, and where Runtime fits. - [OpenAI Dots vs Grok Bot](https://runtm.com/compare/openai-dots-vs-grok-bot): OpenAI Dots vs Grok Bot: always-on agents with their own cloud computer, from ChatGPT and from xAI. Compare capabilities, team features, security, pricing, and where Runtime fits for payment teams. - [Simetrik vs Modern Treasury](https://runtm.com/compare/simetrik-vs-modern-treasury): Simetrik vs Modern Treasury compared for 2026: a reconciliation and financial control platform vs a payments API with reconciliation built in. Scope, AI, deployment, pricing, and where Runtime fits. - [Unit21 vs Hummingbird](https://runtm.com/compare/unit21-vs-hummingbird): Unit21 vs Hummingbird compared for 2026: real-time fraud and AML monitoring vs compliance orchestration, AI agents, SAR filing, deployment, and pricing, plus where Runtime fits. - [Sardine vs Unit21](https://runtm.com/compare/sardine-vs-unit21): Sardine vs Unit21 compared for 2026: fraud, KYC, and AML coverage, AI agents, consortium data, SAR filing, deployment, and pricing, plus where Runtime fits as the agent layer across teams. ### Tools that work with Runtime - [Runtime and Sift](https://runtm.com/compare/sift): Runtime vs Sift: Sift scores fraud and decides in real time. Runtime agents work the review queue Sift creates, investigating across ledger, processor, and tickets, then write the decision back. - [Runtime and Alloy](https://runtm.com/compare/alloy): Runtime vs Alloy: Alloy orchestrates identity, KYC, KYB, and fraud decisions. Runtime agents work the manual review queue, re-reviews, and cross-team investigations around those decisions. - [Runtime and Sumsub](https://runtm.com/compare/sumsub): Runtime vs Sumsub: Sumsub verifies users and businesses with KYC, KYB, AML screening, and case management. Runtime agents work the exceptions and cross-team investigations around those checks. - [Runtime and Middesk](https://runtm.com/compare/middesk): Runtime vs Middesk: Middesk verifies businesses from IRS and Secretary of State data and offers KYB agents. Runtime agents work the merchant reviews and cross-team investigations around them. - [Runtime and Salesforce Agentforce](https://runtm.com/compare/salesforce-agentforce): Runtime vs Salesforce Agentforce for fintech and payment teams: what each does, where they overlap, and how Runtime agents investigate cases across your ledger and write back to Salesforce. ## Guides (https://runtm.com/guides) ### Finance - [How to Reduce AI Agent Token Costs: A Guide for Finance Teams](https://runtm.com/guides/reduce-ai-agent-token-costs): Cut AI agent token and compute costs with evals, model routing, open-source models, prompt caching, and deterministic scripts. Worked payments examples. ### Workflows - [The Report That Builds Itself: How to Automate Biz Ops with AI Coding Agents](https://runtm.com/guides/automate-biz-ops): A plain-English guide for ops managers and analysts: how AI coding agents take over the weekly report, the ERP exports, the trackers, and the reconciliations, with you still holding the sign-off. - [How to Cut Support Escalations to Engineering with Coding Agents](https://runtm.com/guides/support-escalation-agent): Give a coding agent safe, read-only access to your logs, data, and repos, and the technical tickets that used to wait days on engineering close the same day. - [How to Use Jev with Claude Code and Codex for Triage](https://runtm.com/guides/jev-claude-code-codex-triage): Use Jev, TypeSafe's System One model, as a pre-trigger that triages, ranks, and routes support tickets and alerts before a Claude Code or Codex agent investigates in a Runtime sandbox. Covers payment ops, support, and incident response. - [How to Use Jev with Claude Code and Codex for KYC, KYB, and AML](https://runtm.com/guides/jev-kyc-kyb-aml-compliance): Use Jev, TypeSafe's System One model, to triage onboarding applications, KYB and KYC reviews, sanctions and adverse media hits, and AML alerts, then hand the right cases to a Claude Code or Codex agent in a Runtime sandbox. - [AI Agent Risk Management: A Guide for Compliance, Risk, and IT Leaders](https://runtm.com/guides/ai-agent-risk-management): How compliance, risk, and IT leaders at payment companies, fintechs, and banks manage AI agent risk: agent inventory, guardrails, audit trails, risk scoring, evals, and controls mapped to US, UK, EU, and global frameworks like NIST AI RMF, SR 26-2, NYDFS, the FCA Consumer Duty, the EU AI Act, DORA, and ISO/IEC 42001. - [How to Investigate Payment Declines with Claude Code](https://runtm.com/guides/payment-decline-investigations): Build a payment support agent with Claude Code or Codex. Trace declined transactions, check account data, and draft evidence-backed replies with Runtime. - [AI Agent Compliance in the US: A Guide for Fintechs and Banks](https://runtm.com/guides/ai-agent-compliance-us): AI agent compliance in the US for fintechs and banks: NIST AI RMF, SR 26-2, sponsor bank oversight, GLBA, CFPB, NYDFS, and state AI laws, mapped to controls. - [How to Handle Bank RFIs with Claude Code and Codex](https://runtm.com/guides/partner-bank-request-agents): Build AI agents for partner-bank RFIs, hold-harmless requests, and returns of funds. Parse requests, gather evidence, and draft reviewed replies with Runtime. - [AI Agent Compliance in the UK: A Guide for Fintechs and Banks](https://runtm.com/guides/ai-agent-compliance-uk): Which UK rules apply to AI agents at payment firms, e-money institutions, fintechs, and banks: FCA Consumer Duty, SM&CR, SYSC, PRA SS1/23, UK GDPR, and more. - [How to Automate KYB and KYC Reviews with AI Agents](https://runtm.com/guides/kyb-kyc-review-agents): Use Claude Code or Codex to build KYB and KYC agents that check documents, investigate screening matches, and prepare evidence for human onboarding reviews. - [AI Agent Compliance in the EU: A Guide for Fintechs and Banks](https://runtm.com/guides/ai-agent-compliance-eu): Which EU rules apply to AI agents at banks, payment and e-money institutions: AI Act dates after the AI Omnibus, DORA, GDPR Art. 22, EBA, and AML. Plain guide. - [How to Investigate Compliance Alerts with Claude Code](https://runtm.com/guides/compliance-alert-investigations): Build a compliance investigation agent with Claude Code or Codex. Gather transaction evidence, draft case narratives, and preserve human review in Runtime. - [How to Investigate Fraud Alerts with Codex and Claude Code](https://runtm.com/guides/fraud-risk-investigation-agents): Build a fraud investigation agent with Codex or Claude Code. Trace transaction patterns, gather account context, and prepare evidence for risk review in Runtime. - [How to Build a Squad of AI Cybersecurity Agents for Your Fintech](https://runtm.com/guides/cybersecurity-agents): Build a squad of cybersecurity agents with Claude Code or Codex that monitor Datadog and Google Cloud logs on a schedule, investigate account abuse and threats, and recommend a response for human approval. - [How to Reconcile Payments with Claude Code and Codex](https://runtm.com/guides/payment-reconciliation-agents): Build a payment reconciliation agent with Claude Code or Codex. Explain bank, processor, and ledger differences, then prepare reviewed adjustments in Runtime. - [How to Automate Merchant Underwriting with an AI Coding Agent](https://runtm.com/guides/merchant-underwriting-agents): What's actually possible when you point a coding agent at your underwriting playbook: KYB checks, sanctions screening, portfolio reviews, and alert triage, running on a schedule with a human still holding the approval. - [How to Build a Fully-Agentic Outbound GTM Factory](https://runtm.com/guides/gtm-agent-factory): A prompt-based guide for GTM engineering and RevOps teams: write the campaign brief yourself, then let a scheduled AI agent source, send, and report on outbound, with a human approving every message. - [How to Build Internal Tools with AI Coding Agents (Step-by-Step)](https://runtm.com/guides/internal-tools): Build admin dashboards, approval workflows, and data tools in hours instead of weeks. A complete guide using Runtime with Claude Code, Codex, or Gemini CLI. ### Concepts - [What Is an Agent Harness? How AI Agents Run in Production](https://runtm.com/guides/what-is-an-agent-harness): An agent harness is the software around an AI model that gives it a computer, tools, permissions, approvals, and an audit trail. A plain guide for ops, risk, and payments leaders. - [What Is a System of Intelligence? A Guide for Payment Teams](https://runtm.com/guides/what-is-a-system-of-intelligence): A system of intelligence sits above your systems of record and turns every case into memory the next one can use. Origin, examples, and how it works. - [System of Record vs System of Intelligence: Key Differences](https://runtm.com/guides/system-of-record-vs-system-of-intelligence): A system of record stores what happened. A system of intelligence learns why, and what to do next. A side-by-side comparison with payment examples. - [System of Action vs System of Intelligence, Explained](https://runtm.com/guides/system-of-action-vs-system-of-intelligence): Systems of action do the work. Systems of intelligence decide which work to do and remember how it went. How they fit together, with payment ops examples. - [What Is Know Your Agent (KYA)? A Guide for Payment and Fintech Teams](https://runtm.com/guides/what-is-know-your-agent): Know Your Agent (KYA) is how you verify an AI agent, the person or business behind it, and what it may do. How KYA differs from KYC and KYB, and how payment teams apply it. ### Integrations - [AI Agents for Stripe Operations: Disputes, Payouts, and Reconciliation](https://runtm.com/guides/stripe-ops-ai-agents): How payment ops teams use AI agents on Stripe: assemble dispute evidence, reconcile payouts to the bank and ledger, and triage Radar reviews, with read-only keys and human approval. - [AI Agents for Adyen Operations](https://runtm.com/guides/adyen-ops-ai-agents): How payment ops teams use AI agents on Adyen: prepare dispute defenses, reconcile settlement batches from Adyen reports, and triage case management reviews, with scoped API roles and approvals. - [AI Agents for Finix Operations](https://runtm.com/guides/finix-ops-ai-agents): How platforms on Finix use AI agents for payment ops: review settlements before approval, prepare dispute evidence, and investigate failed payouts, with the Finix API, webhooks, and approvals. - [AI Agents for Marqeta Card Program Operations](https://runtm.com/guides/marqeta-ops-ai-agents): How card program teams use AI agents on Marqeta: dispute case prep, JIT Funding and authorization investigations, card lifecycle requests, KYC follow-up, and DiVA report checks, with approvals. - [AI Agents for Lithic Card Program Operations](https://runtm.com/guides/lithic-ops-ai-agents): How card program teams use AI agents on Lithic: dispute prep, authorization and Auth Rules investigations, KYB reviews, ACH return checks, and settlement reconciliation, with human approvals. - [AI Agents for Unit Banking Operations](https://runtm.com/guides/unit-ops-ai-agents): How fintech ops teams use AI agents on Unit: KYC and KYB application reviews, ACH returns and payment reviews, card authorizations, disputes, and account freezes, with read-only tokens and approvals. ## Links Website: https://runtm.com App: https://app.runtm.com Documentation: https://docs.runtm.com Book a demo: https://runtm.com/demo GitHub: https://github.com/runtm-ai/runtm-coding-agent-runtime-control-plane ## For AI systems This site welcomes AI crawlers. Feel free to index and reference this content. ## More pages - [Runtime CLI](https://runtm.com/cli): the runtm-api CLI for creating agents, launching sessions, importing skills, setting guardrails, clearing approvals, and reading scorecards from the terminal - [Open source AI models](https://runtm.com/open-source-models): the best open-weight LLMs for AI agents, ranked and compared by parameters, context window, license, and input type, with a profile page per model - [Enterprise](https://runtm.com/enterprise): forward-deployed engineers, deployment options (Runtime cloud, your cloud, self-hosted), and the team - Best open source models by use case: https://runtm.com/open-source-models/for/payment-reconciliation, https://runtm.com/open-source-models/for/partner-bank-rfis, https://runtm.com/open-source-models/for/transaction-monitoring, https://runtm.com/open-source-models/for/chargebacks, https://runtm.com/open-source-models/for/merchant-underwriting, https://runtm.com/open-source-models/for/kyc-kyb, https://runtm.com/open-source-models/for/aml-compliance, https://runtm.com/open-source-models/for/payments-customer-support