Maywood is an AI software platform that lets investment banks automate their entire dealmaking workflow.

The Maywood platform automatically generates finished, on-brand CIMs / presentations, models, and diligence responses that prioritizes editability and auditability, with best-in-class security ensuring information stays protected.

Maywood reinvents the deal process with a single, living workspace for every deal.

AI Agent Engineer

Maywood AI is building AI agents that transform how teams work with data and presentations—automating complex workflows that currently take hours of manual effort.

What You'll Build

Presentation Automation Agents

• Design agents that generate investor decks, board reports, and client presentations from raw data
• Build systems that understand narrative structure, visual hierarchy, and business context
• Optimize for brand consistency, slide design quality, and content accuracy
• Reduce multi-hour presentation tasks to minutes while maintaining professional polish

Financial Analysis Agents

• Create agents that parse financial statements, extract insights, and generate executive summaries
• Work with large datasets—earnings reports, market data, company filings
• Build reasoning systems that spot trends, anomalies, and key drivers without hallucinating
• Design agent architectures that explain their analysis and show their work

Agent Architecture & Optimization

• Reduce dependence on LLMs through smart prompt engineering and policy design
• Maintain accuracy and consistency across long, multi-step workflows
• Evaluate using real business outcomes: Did the deck close the deal? Was the analysis actionable?
• Ship creative, non-obvious techniques that make agents more reliable and cost-effective

Systems & Deployment

• Own the full lifecycle: prototype → deploy → monitor → iterate
• Build production-grade systems that handle real customer workloads
• Design integrations with data sources, cloud infrastructure, and customer tools
• Set up CI/CD, logging, and quality monitoring for live agents

What We're Looking For

People who can build agents that don't just work—they work well. You care about the details: Does the analysis make sense? Does the presentation flow? Would a human trust this output?