D

doodlestein

Independent analyst and commentator focused on enterprise AI adoption, agent orchestration, and legal/regulatory risks around model training. Offers short, idea-first posts that connect technical details to commercial and investment implications.

Trust score
0 / 100
Track record
0 / 100
Thesis calls
34
Evaluated calls
30
Average return
+10.53%
Win rate
57%

Past bets that played out

Highlights include repeated notes that copyright litigation risk from training data may be manageable because information is "laundered" into model weights, and practical observations about using tools like Opus to surface bugs in long goal‑oriented sessions. Also publishes conceptual notes on agent fungibility and orchestration.

MUrightbacktest HOLD

Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 28 / 100Return: +105.38%
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
STXrightbacktest HOLD

Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 20 / 100Return: +104.12%
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
AMDrightbacktest PROMOTE

The source is a pinned post by Jeffrey Emanuel linking to “The Short Case for Nvidia Stock” (via youtubetranscriptoptimizer.com). No substantive arguments are included in the provided text beyond indicating a bearish/short stance on Nvidia.

Mentioned: Jun 17, 2026, 10:42 PM EDTConviction: 22 / 100Return: +90.55%
Source: Pinned Jeffrey Emanuel @doodlestein Jan 25, 2025 Replying to @doodlestein You can read it here— let me know what you ...

What this channel is watching now

Most frequently discussed tickers: MSFT (3 mentions, avg conviction 0.3267), NVDA (2 mentions, 0.52), ADBE (2 mentions, 0.375), AMZN (2 mentions, 0.35). Coverage centers on enterprise AI platforms, developer risk, and infrastructure for on‑prem and open‑model deployments.

Latest videos and market context

No video content available.

Jeffrey Emanuel @doodlestein 26m One important update to this now that I’ve used it for longer: Opus 5 seems to be mu...

Jul 25, 2026, 10:33 AM EDT

Post compares two AI models/tools (“Opus 5” vs “Fable 5”). Key actionable operational guidance: Opus 5 may suffer “context rot” beyond ~500k tokens; restart sessions to avoid performance degradation. Not directly tied to public companies, revenues, or near-term catalysts; limited market tradability without identifying the vendors behind Opus/Fable.

Jeffrey Emanuel @doodlestein 23s Crazy to see projects getting so much attention that basically do the same thing as ...

Jul 25, 2026, 9:52 AM EDT

Post notes attention on a newly promoted open-source “trajectory” format for coding-agent session/experience data, while the author points out a similar open-source project (“cass”) has existed for 6+ months. This is primarily developer-tooling/AI-agent ecosystem chatter; tradability is indirect via platform beneficiaries (cloud/AI model hosts) and devtools proxies.

Jeffrey Emanuel @doodlestein 44m The thinking man’s pile of losing scratch-and-win tickets (i.e., artifacts left in y...

Jul 25, 2026, 6:02 AM EDT

This post is a humorous, non-financial comment about a long GPT Pro session and a math conjecture. It contains no market-relevant claims, catalysts, sectors, or tradable implications.

Jeffrey Emanuel @doodlestein 17m The Opus 5 mega orchestration prompt if you’re a user of my skills site and my other...

Jul 24, 2026, 9:08 PM EDT

Social post sharing an “Opus 5 mega orchestration prompt” for software/codebase investigation workflows. No explicit company, product launch, financials, adoption metrics, or market-moving information; mostly how-to prompt content.

Proof-backed call history

Publishes short, idea-driven posts and replies connecting technical AI topics to market and legal implications. Recommendations and commentary emphasize pragmatic adoption paths for enterprise AI and the limits of regulatory/legal overhang on commercialization.

IQwrongbacktest DEMOTE

Post compares two AI models/tools (“Opus 5” vs “Fable 5”). Key actionable operational guidance: Opus 5 may suffer “context rot” beyond ~500k tokens; restart sessions to avoid performance degradation. Not directly tied to public companies, revenues, or near-term catalysts; limited market tradability without identifying the vendors behind Opus/Fable.

Mentioned: Jul 25, 2026, 10:33 AM EDTConviction: 60 / 100Return: -2.88%
Source: Jeffrey Emanuel @doodlestein 26m One important update to this now that I’ve used it for longer: Opus 5 seems to be mu...
OPENAIopen

...his case, cass: github.com/Dicklesworthst… Letta @Letta_AI Jul 24 Article Trajectory: A Standard Format for Agent Experience Data Introducing trajectory, an open-source package that normalizes coding-agent sessions from @AnthropicAI Claude Code, @OpenAI Codex, @pidotdev, @LangChain deepagents, @openclaw, Letta Code, and other... 1 27 Jeffrey Emanuel @doodlestein 23s Crazy to see projects getting so much attention that basically do the same thing as my own projects that have been out for 6+ mo

Mentioned: Jul 25, 2026, 9:52 AM EDTConviction: 100 / 100
Source: Jeffrey Emanuel @doodlestein 23s Crazy to see projects getting so much attention that basically do the same thing as ...
ANTHROPICopen

...+ months. In this case, cass: github.com/Dicklesworthst… Letta @Letta_AI Jul 24 Article Trajectory: A Standard Format for Agent Experience Data Introducing trajectory, an open-source package that normalizes coding-agent sessions from @AnthropicAI Claude Code, @OpenAI Codex, @pidotdev, @LangChain deepagents, @openclaw, Letta Code, and other... 1 27 Jeffrey Emanuel @doodlestein 23s Crazy to see projects getting so much attention that basically do the same thing as my own projects that have been

Mentioned: Jul 25, 2026, 9:52 AM EDTConviction: 100 / 100
Source: Jeffrey Emanuel @doodlestein 23s Crazy to see projects getting so much attention that basically do the same thing as ...
ANTHROPICopen

...e agent I have going with 3 Opus 5 agents. Claude @claudeai 1h Introducing Claude Opus 5. It's a thoughtful and proactive model that comes close to the frontier intelligence of Fable 5 at half the price. 00:08 9 1 40 2.4K Social post highlights Anthropic announcing Claude Opus 5: near “frontier” quality relative to “Fable 5” at ~half the price. A user plans to replace each existing “Fable agent” with multiple Opus 5 agents, implying a step-change in cost/performance that could accelerate AI a

Mentioned: Jul 24, 2026, 1:15 PM EDTConviction: 100 / 100
Source: Jeffrey Emanuel @doodlestein 53m Brb, gonna replace each Fable agent I have going with 3 Opus 5 agents. Claude @claud...
OPENAIopen

...for the same thing. It sucks because I noticed those log files growing uncontrollably and should've stopped it. Uzi @uzairansar Jun 22 So Codex has a bug that could kill your SSD in under a year... 48 29 788 157K Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 100 / 100
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
MUrightbacktest HOLD

Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 28 / 100Return: +105.38%
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
STXrightbacktest HOLD

Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 20 / 100Return: +104.12%
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
WDCrightbacktest HOLD

Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 25 / 100Return: +86.14%
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
SSNLFwrongbacktest HOLD

Anecdotal social post alleging an OpenAI Codex-related bug caused runaway log-file writes that killed a Samsung 9100 PRO 4TB PCIe 5.0 SSD in ~13 months; highlights potential software-induced write amplification risk and notes a sharp price increase for the same SSD model.

Mentioned: Jul 23, 2026, 8:50 PM EDTConviction: 22 / 100Return: 0.00%
Source: Jeffrey Emanuel @doodlestein 8h Damn, looks like this Codex bug really did kill my SSD, which was only 13 months old ...
GOOGLrightbacktest HOLD

Discussion claims Google Search revenue growth is being driven by an “artificial” increase in ad costs (possibly from auction mechanics/customer “milking”), implying recent strong results may be less durable. However, reader-added context disputes the key mechanism: Google Search still uses a second-price auction; first-price changes were primarily for display ads. Net: an anecdotal bearish narrative on Google’s ads quality/durability, partially undercut by factual correction.

Mentioned: Jul 23, 2026, 4:47 PM EDTConviction: 43 / 100Return: +21.59%Observed price: $317.69
Source: Jeffrey Emanuel @doodlestein 12h If this is true (and it sounds true to me, especially if they really deprecated the ...
GOOGLrightbacktest DEMOTE

Discussion claims Google Search revenue growth is being driven by an “artificial” increase in ad costs (possibly from auction mechanics/customer “milking”), implying recent strong results may be less durable. However, reader-added context disputes the key mechanism: Google Search still uses a second-price auction; first-price changes were primarily for display ads. Net: an anecdotal bearish narrative on Google’s ads quality/durability, partially undercut by factual correction.

Mentioned: Jul 23, 2026, 4:47 PM EDTConviction: 43 / 100Return: -21.59%Observed price: $317.69
Source: Jeffrey Emanuel @doodlestein 12h If this is true (and it sounds true to me, especially if they really deprecated the ...
AVGOrightbacktest PROMOTE

The source is a pinned post by Jeffrey Emanuel linking to “The Short Case for Nvidia Stock” (via youtubetranscriptoptimizer.com). No substantive arguments are included in the provided text beyond indicating a bearish/short stance on Nvidia.

Mentioned: Jun 17, 2026, 10:42 PM EDTConviction: 18 / 100Return: +23.53%
Source: Pinned Jeffrey Emanuel @doodlestein Jan 25, 2025 Replying to @doodlestein You can read it here— let me know what you ...

About this channel

doodlestein is a platform commentator and analyst on X (@doodlestein) who writes briefly but analytically about AI model training risks, enterprise adoption patterns, and agent orchestration. Work tends to be conceptual with occasional practical observations about tooling and workflows.

Subscribersn/a
Videosn/a
Win rate57%
Average return+10.53%

@doodlestein

Unlock the full track record

Follow @doodlestein for concise takes on enterprise and generative AI, model-training legal risk, and practical notes about running production AI workflows on open or on‑prem stacks.

22 more thesis calls are available after sign-up.