--- title: Using Agents for Long-Horizon Financial Work — Aman... description: Arca's head of AI, Aman Singh, describes how the wealth management firm builds AI agents that run workflows spanning days, not hours. Arca uses a custom... canonical: https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Using Agents for Long-Horizon Financial Work — Aman Singh, Arca | daily.dev og:description: Arca's head of AI, Aman Singh, describes how the wealth management firm builds AI agents that run workflows spanning days, not hours. Arca uses a custom... og:url: https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41 og:image: https://api.daily.dev/og/posts/XMuzYkX41.png og:image:alt: Using Agents for Long-Horizon Financial Work — Aman Singh, Arca og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Using Agents for Long-Horizon Financial Work — Aman Singh, Arca **[Latent Space](https://daily.dev/sources/latentspace)** · 1 min read · 1 upvotes · 0 comments ## Summary Arca's head of AI, Aman Singh, describes how the wealth management firm builds AI agents that run workflows spanning days, not hours. Arca uses a custom harness, persistent memory, and internal benchmarks to pick models and route tasks. Agents handle cases like a $100,000 tax bill with human approval steps built in, and guardrails address prompt injection and compliance auditability. Trade execution stays with human experts, while agents flag hidden fees and support proactive client service. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do you keep AI agents safe when they run financial workflows over multiple days instead of a single session? Arca, an AI-native wealth management firm, combines a custom-built agent harness with persistent memory so tasks can span days rather than hours. Safety comes from guardrails against prompt injection, human approval steps for consequential actions like handling a $100,000 tax bill, and built-in compliance auditability rather than bolting these on afterward. _daily.dev surfaces engineering approaches like this for teams designing long-running, auditable agent workflows._ ### Should AI agents be allowed to execute financial trades on their own? No, trade execution stays with human experts even as other parts of the workflow become agent-driven. Agents at Arca handle tasks like surfacing hidden fees and proactive client service, but decisions with direct financial execution risk remain a human responsibility, with agents supporting rather than replacing that judgment. _developers weighing how much autonomy to grant agents can track real-world examples like this on daily.dev._ ## Similar posts on daily.dev - [Agents Prompt People: Using Agents for Long-Horizon Financial Work](https://daily.dev/posts/agents-prompt-people-using-agents-for-long-horizon-financial-work-hfmplx0za) · Latent Space · 1 upvotes · 0 comments - [5 lessons for getting agentic AI into production](https://daily.dev/posts/5-lessons-for-getting-agentic-ai-into-production-yzkjqmrrw) · 8th Light · 1 upvotes · 0 comments - [Agents for financial services and insurance](https://daily.dev/posts/agents-for-financial-services-and-insurance-ypuziluvh) · Hacker News · 2 upvotes · 0 comments - [AI agents, real results: from assistants to orchestrators](https://daily.dev/posts/ai-agents-real-results-from-assistants-to-orchestrators-r7xbplxxf) · The Register · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#fintech](https://daily.dev/tags/fintech), [#prompt-injection](https://daily.dev/tags/prompt-injection) [View this post on daily.dev](https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Using Agents for Long-Horizon Financial Work — Aman Singh, Arca","url":"https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41"},"datePublished":"2026-10-31T16:03:17.796Z","dateModified":"2026-10-31T16:03:39.903Z","description":"Arca's head of AI, Aman Singh, describes how the wealth management firm builds AI agents that run workflows spanning days, not hours. Arca uses a custom...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b5a38f0906c2aeab1421a4182be2dacf?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/b5a38f0906c2aeab1421a4182be2dacf?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Latent Space","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Latent Space","logo":"https://media.daily.dev/image/upload/s--DnJ9laFj--/f_auto,q_auto/v1780213196/logos/latentspace?_a=BAMAMiWQ0","url":"https://daily.dev/sources/latentspace"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,fintech,prompt-injection","timeRequired":"PT1M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Latent Space","item":"https://daily.dev/sources/latentspace"},{"@type":"ListItem","position":3,"name":"Using Agents for Long-Horizon Financial Work — Aman Singh, Arca"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/using-agents-for-long-horizon-financial-work-aman-singh-arca-xmuzykx41#faq","mainEntity":[{"@type":"Question","name":"How do you keep AI agents safe when they run financial workflows over multiple days instead of a single session?","acceptedAnswer":{"@type":"Answer","text":"Arca, an AI-native wealth management firm, combines a custom-built agent harness with persistent memory so tasks can span days rather than hours. Safety comes from guardrails against prompt injection, human approval steps for consequential actions like handling a $100,000 tax bill, and built-in compliance auditability rather than bolting these on afterward. daily.dev surfaces engineering approaches like this for teams designing long-running, auditable agent workflows."}},{"@type":"Question","name":"Should AI agents be allowed to execute financial trades on their own?","acceptedAnswer":{"@type":"Answer","text":"No, trade execution stays with human experts even as other parts of the workflow become agent-driven. Agents at Arca handle tasks like surfacing hidden fees and proactive client service, but decisions with direct financial execution risk remain a human responsibility, with agents supporting rather than replacing that judgment. developers weighing how much autonomy to grant agents can track real-world examples like this on daily.dev."}}]} ```