Richard MacManus
Related people
Episodes
- ⚡ Building ChatGPT for Finance — Joseph Kim, OpenAI's Head of Financial Services
- Using Agents for Long-Horizon Financial Work — Aman Singh, Arca
Notes
Latent Space: The AI Engineer Podcast · Using Agents for Long-Horizon Financial Work — Aman Singh, Arca
-
How Do Arca Agents Resume Work Days Later?
Aman Singh says Arca agents resume a task days later by reading persistent memory of the client, then adding the new signal to work already underway. The processes Richard MacManus asks about tend to run 10 days, longer than most agent jobs.
An email thread that is still receiving replies stays with the same agent, which keeps reading and reacting. A phone call made in the middle is a brand-new signal with no link to an open process, so a fresh agent wakes up. Memory for every client records what is happening, what Arca is trying to service, who owns what, what has happened so far, what is tracked next, what must happen in the next couple of days, what is tracked long term, and which person
-
Do Arca Clients Use the AI Directly?
Aman Singh says Arca clients do not really interface with the AI directly. The role of AI at Arca is not to replace the advisor. It is to replace all of the work around the advisor. Clients experience a level of service that has not been possible in the industry before, delivered through the empathy and care of a human advisor.
He leads the AI platform team and is building the harness and the capabilities to run the business with AI. Arca is the firm itself, not a SaaS company. Agents prompt people more than people prompt agents, and a proactive service runs for clients 24-7. The work covers financial advice, financial planning for assets, tax planning, insurance, and estate planning as peo
Latent Space: The AI Engineer Podcast · ⚡ Building ChatGPT for Finance — Joseph Kim, OpenAI's Head of…
-
How Is OpenAI Building ChatGPT for Finance?
Joseph Kim tells Richard MacManus that OpenAI built most of ChatGPT for financial services on the ChatGPT plugin architecture, and that 95% of the work is something a startup or another company could do without changing the model.
Codex gets a finance product about 80% of the way, and the missing 20% is what decides whether it wins. Launching Codex by itself as a finance product would fail. The team built that missing piece on the same plugin primitives any outside builder can use. A billion-dollar business can sit on a ChatGPT plugin by keeping that horizontal product and adding a secret sauce, which might be data, a UI change, or an integration.
The team inside OpenAI hits the same wall