The AI Daily Brief: "The Coolest Agents I've Built So Far"

Host: Nathaniel Whittemore (NLW)
Date: March 14, 2026
Theme: A playful "March Madness"-style tournament sharing and ranking the 16 coolest AI agents and projects Nathaniel has built so far this year.

Episode Overview

In this behind-the-scenes bonus episode, Nathaniel Whittemore (“NLW”) hosts his own “Agent Madness”—a single-elimination bracket pitting 16 AI agents/projects he developed this year against each other. Each match-up is judged on technical complexity, real-life usefulness, value beyond Nathaniel’s personal workflows, and a healthy dose of X-factor. The episode offers not just a showcase of current agent technology but also insight into where agentic tooling is headed, both for individuals and organizations.

Nathaniel’s tone is enthusiastic, candid, and sometimes playfully self-deprecating. The episode is structured around the tournament, with quick summaries of each agent, pros and cons, and justification for his choices as they advance through the bracket.

Key Discussion Points & Notable Projects

The Agentic Shift in AI

• Opening Context:
  • NLW remarks on the explosion in agent-building in the past several months—“there has been a massive shift over the last three to four months. It is an agentic shift…” [00:42]
  • Mentions tools/platforms enabling this shift: OpenClaw, Claude Code, Codecs, Perplexity Computer.
  • Community angle: Many listeners are participating/building/sharing, referencing programs like AIDB New Year's, CLAW Camp, and Enterprise Claw.

The Agent Madness Tournament Structure

• Personalized Brackets:
  • 16 projects, seeded by both Claude and ChatGPT (“they actually came up with almost exactly the same seeding.” [02:11])
  • No theme-based divisions; brackets created to ensure diverse and competitive match-ups.
  • Criteria: technical complexity, utility, broader value, gut feel or “X factor.”

Bracket A - Project Matchups

1. Holmes Agent vs. AI Daily Brief Website

• Website:
  • Simple, functional Lovable-based podcast hub.
  • “Ranks about as low as it gets on technical complexity. After all, it's just a website built with lovable…” [03:00]
• Holmes Agent:
  • Named for Sherlock Holmes, part of a broader agent ecosystem for AI strategy.
  • Individual-focused recommendations; collects detailed “case files” on users and suggests tailored AI workflow improvements.
  • “Holmes has a web interface where he can interview you about your work... and he also has a presence in Slack...” [04:53]
• Winner: Holmes Agent
  • “Although I do have that fondness for aidailybrief AI, obviously I'm going to give this one to Holmes.” [06:36]

2. OpenClaw Coder/Widi Builder vs. Perplexity Research Library

• OpenClaw Coder (“Widi Builder”):
  • Coder bot controllable via Telegram.
  • Limited long-term utility after Claude Code introduced remote features.
  • “Ultimately, this did not really enter into my rotation…” [07:11]
• Perplexity Research Library:
  • Automated AI research aggregator.
  • Promise as a knowledge management tool, but weak on generative, natural language search.
• Winner: OpenClaw Coder
  • Edges out due to newer iterations and database integration. [08:03]

3. Chucky vs. AIDB New Year

• Chucky:
  • Inspired by Good Will Hunting—agent as a representative for an AI builder.
  • Demonstrates builder’s actual skills interactively, not just through resumes or portfolios.
  • “Chucky is, is an agent that acts as an agent builder's representative... you send them your agent representative, Chucky, who can interact with them and tell them about what you've built.” [09:11]
• AIDB New Year:
  • 10-week self-directed learning program; over 7,000 participants.
  • Significant community impact and inspiration for later projects.
• Winner: Chucky
  • For future potential as a new work-credentialing paradigm. [12:24]

4. 221B vs. Model MOG

• 221B:
  • “The brain” for Holmes and Mycroft—an agentic knowledge hub.
  • Aggregates show transcripts, web searches, and interviews; feeds updates to other agents.
• Model MOG:
  • Fun prototype to let users express model preferences for different use cases.
  • “Model MOG brings up a use case... and asks you between two different options which one you would use.” [14:57]
• Winner: 221B
  • Chosen for centrality to larger agent ecosystems. [16:31]

Bracket B - Project Matchups

1. Claw Camp vs. Mycroft

• Claw Camp:
  • Self-directed agent-building program.
• Mycroft:
  • Holmes’s “older brother” in this fictional universe—an agent for company AI strategy.
  • “If Holmes is concerned just with what the individual is doing with AI, Mycroft is building an overall company strategy.” [18:38]
  • Continuously updates corporate AI roadmaps, collects ongoing organizational data.
• Winner: Mycroft
  • Significant future potential as a “digital chief AI officer.” [20:41]

2. OpenClaw Chief of Staff vs. Mission Control Center

• Chief of Staff:
  • Project manager agent, not fully integrated yet.
  • “Project managers and the Chief of Staff are just kind of externalized to do lists…” [22:34]
• Mission Control:
  • Advanced dashboard managing the expanding swarm of agents.
  • Technically difficult to build; immensely useful.
• Winner: Mission Control Center
  • For complexity and central importance. [23:36]

3. Widi 24/7 Researcher Radar vs. Maturity Maps

• Widi Researcher Radars:
  • Persistent, autonomous research bots looking for AI use cases, studies, tiering them (prime time, emerging, frontier).
  • “Persistent research has been my most useful application of OpenClaw overall.” [25:11]
• Maturity Maps:
  • Visualization/benchmark for organizational AI readiness across six vectors.
• Winner: Widi Researcher Radars
  • Most directly useful currently. [27:06]

4. Agent Madness Platform vs. Beta Superintelligent Compass

• Agent Madness Platform:
  • Bracket management interface for this contest.
• Compass:
  • Enterprise power-tool integrating opportunity radars, maturity maps, self-directed assessments.
• Winner: Compass
  • For vision and integrative features. [28:53]

Semifinals & Finals

• Holmes vs. OpenClaw Coder: Winner – Holmes [29:55]
• Chucky vs. 221B: Winner – Chucky (“Chucky's approach to presenting someone's build history is something that I feel like could be a really valuable form factor going forward.” [30:54])
• Mycroft vs. Mission Control: Winner – Mycroft (“...the idea of a digital chief AI officer just really has legs.” [32:13])
• Widi Researcher vs. Compass: Winner – Widi Researcher (“Compass is evolving...it’s kind of just going to be the enterprise house...” [33:11])

Bracket Finals

• Holmes vs. Chucky: Winner – Chucky (“I think Chucky might be a form factor for the future in terms of presenting the work that you’ve done.” [35:04])
• Mycroft vs. Widi Researcher: Winner – Mycroft (“...doing something valuable with it in a customized and ongoing way, building out the strategy for your company over time.” [36:00])

Agent Madness Champion

• CHAMPION: Mycroft
  • “Mycroft, meanwhile, while it is in testing, is something that I am very excited to release soon and I think might be the best way that I've found so far to scale how I and the teams around me help people figure out AI. So Mycroft is the champion of this Agent Madness AI Mini tournament.” [37:47]

Notable Quotes & Memorable Moments

• On the Agentic Explosion:
  • “There has been a massive shift over the last three to four months. It is an agentic shift...” [00:42]
• On Personal Utility vs. Community Impact:
  • “Not all of them have been fully formed, not all of them have been all that useful, but they've all been, if nothing else, helpful in learning...” [01:44]
• Chucky’s Potential:
  • “I think it could be a really cool way to do jobs matching in the future.” [11:25]
• Why Mycroft Wins:
  • “...the idea of a digital chief AI officer just really has legs.” [32:13]
  • “...might be the best way that I've found so far to scale how I and the teams around me help people figure out AI.” [37:51]
• On Agentic Tool Evolution:
  • “Compass continues to evolve, but I think when it comes to the tournament of the coolest things that I’ve built this year, Compass definitely gets the nod.” [28:58]

Key Timestamps

• 00:42 — Introduction to the "agentic shift" and motivation for Agent Madness.
• 03:00 — Holmes Agent vs AIDB website.
• 07:11 — OpenClaw Coder vs Perplexity Research Library.
• 09:11 — Inspiration and design for Chucky agent.
• 14:57 — Model MOG’s unique approach to model evaluation.
• 18:38 — Mycroft: “digital chief AI officer”.
• 23:36 — Technical challenges of the Mission Control Center.
• 25:11 — Value of persistent research agents.
• 28:53 — Compass as the evolving enterprise hub.
• 35:04 — Why Chucky advances in the finals.
• 37:47 — Mycroft announced as Agent Madness champion.

Conclusion

Through this unique tournament, Nathaniel spotlights the rapid evolution of agentic tooling—highlighting how both individual productivity bots (like Holmes and Chucky) and large-scale organizational agents (like Mycroft and Compass) are shaping the work and learning landscape. Mycroft emerges as the most exciting and scalable agent, poised to democratize AI strategy and roadmap planning at company scale. Light-hearted, insightful, and packed with behind-the-scenes nuggets, the episode is both a celebration and a roadmap for where agent-powered AI is headed next.