Gemini 3.1 Pro is now at the top of the APEX-Agents leaderboard. Gemini jumped from 18.4% to 33.5% on Pass@1 in just 90 days. It also completes 5 tasks that no model has ever been able to do before.…
710387427 viral
AI Agents Benchmarking
Back to Discover Heroes
Brendan Foody
LinkedIn Content Strategy & Writing Style
CEO @ Mercor | Thiel Fellow
United States
79 Viral ScoreView on LinkedIn
1 person tracking this creator on Viral Brain
Performance Overview
Followers
54.7K
Connections
3.7K
Avg. Engagement
Top Posts by Engagement
Posting Patterns & Frequency
1.0 posts/week
Posts / Week
8.1 days
Days Between Posts
1
Total Posts Analyzed
MEDIUM
Posting Frequency
Timezone: UTC
Best Performing Days
ThursdayTuesday
Best Performing Times To Post
17:00–19:00 UTC21:00–23:30 UTC
Topics & Content Focus
Primary Topics
Agentic AI performance benchmarking and model progress narrativesEconomic displacement/augmentation thesis for elite knowledge work (consulting, banking, law) via AI agentsPlatform strategy and network effects framing (building a global expertise/knowledge network)
Secondary Themes
Evaluation infrastructure as the core bottleneck (benchmarks/evals measuring economic value)Reinforcement learning and rapid benchmark saturation dynamicsSocial proof through operator shoutouts and ecosystem credit-giving
Industry Focus
Frontier LLMs applied to enterprise knowledge-work automation (legal deliverables, slide decks, spreadsheet/document workflows)AI agents evaluation/leaderboards (APEX-Agents) as go-to-market narrative infrastructureTalent/expertise marketplaces evolving into knowledge networks (Mercor platform positioning)
Content Categories
Benchmark update / leaderboard commentaryStrategic market thesis + predictionCompany narrative positioning (network effects, category creation)Community building / team-alumni recognition
Performance Insights
417.5714285714286%
Avg Engagement Rate
STABLE
Performance Trend
Best Performing Topics
APEX-Agents leaderboard updates with specific metric jumps and capability examplesBroad-scope predictions about AI surpassing professional servicesNetwork-effect platform thesis positioned as a new category (knowledge network)
Virality Signals
Shares spike when posts combine hard metrics + clear, memorizable implication ("better than best consulting firm")Benchmark/leaderboard framing creates 'reference utility' (people share as evidence in debates)Specific deltas over short time windows (90 days, 3 months) increase repost-worthiness
Structure & Quality
95
Avg Length (Words)
HIGH
Depth Level
ADVANCED
Expertise Level
0.78/10
Uniqueness Score
Common Hooks
Leaderboard/benchmark superlative hook ("best we've ever tested", "now at the top")Compressed time-horizon contrast ("a year ago... now...") to signal accelerationHard threshold milestone ("first to pass 50%") as credibility anchorCategory-level claim framed as inevitability ("imminently", "most important network effect")Named-entity credibility stacking (models, benchmarks, products, people)
Common Endings
Future-forward conclusion that reframes the implication at economy scaleAttribution/shoutout ending ("great work", naming individuals/teams)Congratulatory close to a major lab/company releaseShort community tag/identity marker (team/alumni signal)
Value Delivery Methods
Turns model eval results into executive-level implications (market timing + category impact)Provides concrete progress signals (thresholds, deltas, task capability examples) that reduce ambiguityClarifies the real bottleneck (measurement/evals) rather than repeating generic 'AI is improving' claimsBuilds trust through credit-giving and ecosystem awareness
Formatting Style
Short paragraphs with line breaks for scanabilityMetric-first statements followed by implication framingMinimal emoji usage (used sparingly as identity/energy marker)Proper nouns and technical terms used without over-explaining (assumes informed audience)
Audience & Tone
NO
Question Usage
0.12%
Response Rate
Detected Tone
Benchmark-driven futurist-operatorClinical credibility with occasional insider-casual signalingConfident, thesis-forward, economy-scale framingHigh-agency builder narrative (platform/network effects orientation)semi-formalfirst-person
Interaction Style
Debate-seeking through bold predictions (invites rebuttal and scenario-building)Peer validation loops (tagging/crediting practitioners and labs)Evidence-based discussion starter (metrics as shared ground for comments)
Community Building Signals
In-group identity markers (e.g., team/alumni 'mafia' language)Public recognition of individuals/teams to strengthen network tiesPositions company narrative within a larger movement (agents transforming knowledge work)
Writing Style Patterns
Content Strategy
Hook: Leaderboard/benchmark superlative hook ("best we've ever tesTone: semi-formalCTA: Implicit CTA via strong claims designed to trigger
Writing style breakdown
<start of post>
The speed at which LLMs are mastering specialized labor is breathtaking. GPT 5.4 just hit a 62% success rate on the APEX-Agents legal reasoning module.
Six months ago, no model could maintain consistency across a 50-page contract. Now, GPT 5.4 identifies conflicting clauses and suggests redlines in seconds.
This is the end of rote document review. The bottleneck is no longer human hours, but the quality of the underlying data.
The economic impact on professional services will be enormous.
great work Sam Altman and the OpenAI team
<end of post>
Sign in to unlock the full writing analysis
Nail your LinkedIn strategy with ViralBrain.
Analyze and write in Brendan Foody's style. Grow your LinkedIn to the next level.
