# Your Finance Agent's Bottleneck Is You — Ramana Siddanth Emani, Auditoria AI

## Executive summary

The primary bottleneck in developing production AI agents, especially for regulated fields like finance, is not model capability or hardware, but the developer's manual workflow. The talk outlines how advanced agent harnesses can automate and multiply human output by implementing primitives such as sub-agents, skills, goals, and loops. By orchestrating these agents across parallel worktrees (e.g., handling multiple Jira tickets simultaneously), developers can drastically reduce time spent on repetitive tasks like testing, building, and deployment, shifting the human role from executor to final verifier.

## Key takeaways

- The Developer Loop is the Bottleneck: Shipping production agents fails due to developer velocity limitations. The solution lies in automating the internal developer loop rather than waiting for model or hardware upgrades (0:00, 2:37).
- Parallel Task Execution via Sub-Agents: Sub-agents allow for parallel processing using isolated 'worktrees'—virtual folders where agents write and execute code independently. This enables handling multiple tasks (e.g., QA bug tickets) simultaneously, far exceeding human capacity (4:22).
- Minimal UX for Orchestration: Effective agent use requires minimal User Experience (UX), consolidating all project dashboards, logs, Jira tickets, and GitHub PRs into a single pane of glass widget to reduce context switching and manual orchestration effort (5:18).
- Achieving Autonomy with Goals and Dreaming: Future agent development involves combining 'goals' (setting a desired outcome) with automated loops. Furthermore, allowing agents to 'dream' by collecting customer usage patterns can drive self-improvement and feature upgrades autonomously (11:35).

## Technical details

- Agent Primitives: The core mechanisms for automation include: **Sub-agents** (for parallel task execution), **Skills** (organizational best practices/workflows that guide agents to use correct patterns), and **Goals/Loops** (setting high-level objectives for autonomous problem solving).
- Development Workflow Automation: The automated process flow can include: parsing requirements, root cause analysis, pulling traces and logs, performing Test Driven Development (TDD), implementing fixes, creating a Pull Request (PR), local end-to-end testing, building Docker images, and deploying to staging/development environments.
- System Architecture: The system must integrate various sources of data—logging systems, authentication gateways, Jira tickets, GitHub PRs, and cloud code sessions—into a single minimal UX dashboard.

## Practical implications

- Developers should shift focus from manual execution (writing code, running tests) to defining high-level goals and verifying agent outputs.
- The concept of 'Skills' must be formalized within an organization to ensure agents follow compliant and correct workflows, especially in finance.
- Adopting a minimal UX dashboard is critical for orchestrating multiple sub-agents and monitoring the entire development lifecycle from one interface.

## Topics

AI Agents, DevOps Automation, Build Engineering, Finance Technology, Software Development Life Cycle (SDLC), Auditoria AI, Siddanth Emani (Speaker)

Source: https://www.youtube.com/watch?v=z0sh8HyTrDo