jukan05
Market commentary and short-form analysis on AI, semiconductors, and tech infrastructure. Practical, thematic takes on capex cycles, supply-chain execution, and talent-driven moats.
Past bets that played out
Notable calls include a capex-cycle thesis that Meta may invest ~$600B in AI infrastructure by 2028—framed as a datacenter supply-chain catalyst and a margin/FCF risk for META—and a persistent execution-advantage argument for TSMC based on 24/7 PhD-level fab work culture that is hard to replicate in the U.S.
Post claims Zuckerberg said Meta plans to invest ~$600B in AI infrastructure by 2028, with already-guided CapEx of ~$70B (2025) and ~$100B (2026), implying a sharp ramp to ~$200B (2027) and ~$300B (2028) to hit $600B total. Actionable primarily as a capex-cycle catalyst for AI datacenter supply chain beneficiaries and a margin/FCF risk for META if spend ramps as implied.
Post argues Taiwan/TSMC’s 24/7 fab work culture (PhD engineers on rotating shifts) is structurally difficult to replicate in the U.S., implying sustained execution advantage for TSMC and skepticism toward U.S. onshoring “semiconductor dream” narratives. Mentions $INTC as culturally less aligned with the required intensity.
Post claims Zuckerberg said Meta plans to invest ~$600B in AI infrastructure by 2028, with already-guided CapEx of ~$70B (2025) and ~$100B (2026), implying a sharp ramp to ~$200B (2027) and ~$300B (2028) to hit $600B total. Actionable primarily as a capex-cycle catalyst for AI datacenter supply chain beneficiaries and a margin/FCF risk for META if spend ramps as implied.
What this channel is watching now
Frequent commentary centers on NVIDIA (NVDA) and Meta (META), with additional emphasis on xAI (XAI) and TSMC (TSM). Themes: AI infrastructure demand, compute/talent implications, and structural advantages in semiconductor manufacturing.
Latest videos and market context
Recent YouTube posts from this source. Create an account to unlock live alerts and the full research trail.
Jukan @jukan05 Oct 20, 2025 Honestly, DeepSeek trades better than I do. A 42% return… that’s insane. How many more ye...
Post remarks that “DeepSeek” achieved a 42% return and muses that AI/quant systems may replace human hedge-fund traders within ~10 years. No cashtags or public companies mentioned; “DeepSeek” is not clearly a tradable public-market ticker in this text. Actionability is low because it’s a broad narrative without a catalyst, positioning, or investable instruments referenced.
Jukan @jukan05 Jul 3, 2025 Z Fold 7
Post contains only the phrase “Z Fold 7” with no additional context, claims, or market-relevant details. It is not sufficiently actionable to derive a supported investable implication or ticker-specific trade idea.
Jukan @jukan05 Apr 19, 2025 My Thoughts on China Just a few years ago, like most Koreans, I was intensely anti‑China—...
The provided text is a partial/abridged social post thread about personal views on China/anti-China sentiment and a reply about being “pro-CCP”/“Russian propaganda.” It contains no explicit public-market tickers, cashtags, sectors, tradeable assets, macro variables (rates/inflation/oil), policy actions, or company/product-cycle/supply-chain statements that would support an investable implication.
Jukan @jukan05 Oct 18, 2025 At TSMC, PhD-level engineers work in three rotating shifts. The fabs in Taiwan run 24 hou...
Post argues Taiwan/TSMC’s 24/7 fab work culture (PhD engineers on rotating shifts) is structurally difficult to replicate in the U.S., implying sustained execution advantage for TSMC and skepticism toward U.S. onshoring “semiconductor dream” narratives. Mentions $INTC as culturally less aligned with the required intensity.
Proof-backed call history
Author has published 20 tracked recommendations with an evaluated sample of 19, a 73.68% win rate, and average return of 16.9949%. Content mixes short observational posts, thematic sourcing from media (e.g., FT/Reuters), and opinion threads on geopolitics and product cycles.
Post remarks that “DeepSeek” achieved a 42% return and muses that AI/quant systems may replace human hedge-fund traders within ~10 years. No cashtags or public companies mentioned; “DeepSeek” is not clearly a tradable public-market ticker in this text. Actionability is low because it’s a broad narrative without a catalyst, positioning, or investable instruments referenced.
Post remarks that “DeepSeek” achieved a 42% return and muses that AI/quant systems may replace human hedge-fund traders within ~10 years. No cashtags or public companies mentioned; “DeepSeek” is not clearly a tradable public-market ticker in this text. Actionability is low because it’s a broad narrative without a catalyst, positioning, or investable instruments referenced.
Post contains only the phrase “Z Fold 7” with no additional context, claims, or market-relevant details. It is not sufficiently actionable to derive a supported investable implication or ticker-specific trade idea.
Post argues Taiwan/TSMC’s 24/7 fab work culture (PhD engineers on rotating shifts) is structurally difficult to replicate in the U.S., implying sustained execution advantage for TSMC and skepticism toward U.S. onshoring “semiconductor dream” narratives. Mentions $INTC as culturally less aligned with the required intensity.
Post argues Taiwan/TSMC’s 24/7 fab work culture (PhD engineers on rotating shifts) is structurally difficult to replicate in the U.S., implying sustained execution advantage for TSMC and skepticism toward U.S. onshoring “semiconductor dream” narratives. Mentions $INTC as culturally less aligned with the required intensity.
Single short post citing a Reuters framing (“AI is killing consulting”) and highlighting pricing pressure vs in-house cost, explicitly naming Accenture as an example of consultants offering the same service for much less. Actionable mainly as a sector/ticker-level margin/commoditization thesis for IT consulting, but no explicit trade, timing catalyst, or position language from the speaker.
Post speculates that Yann LeCun is leaving Meta, framing it as a pride/reporting-structure issue. No cashtags, no sourcing, and no explicit trading call. If true, it implies potential key-talent/AI-leadership risk for Meta, but the content is rumor/opinion-level and not a confirmed catalyst on its own.
Jukan @jukan05 Oct 12, 2025 FT: Elon Musk’s xAI is developing a technology known as “world models”, which are artific... Jukan @jukan05 Oct 12, 2025 FT: Elon Musk’s xAI is developing a technology known as “world models”, which are artificial intelligence (AI) systems capable of understanding an...
Post relays an FT report that Elon Musk’s xAI is developing “world models” and has hired former Nvidia specialists. This is mostly thematic AI/robotics compute/talent demand evidence, not a direct tradable catalyst. Public-market implications are indirect (AI infrastructure beneficiaries; potential longer-term TSLA/xAI ecosystem optionality), with high uncertainty.
Post relays an FT report that Elon Musk’s xAI is developing “world models” and has hired former Nvidia specialists. This is mostly thematic AI/robotics compute/talent demand evidence, not a direct tradable catalyst. Public-market implications are indirect (AI infrastructure beneficiaries; potential longer-term TSLA/xAI ecosystem optionality), with high uncertainty.
Very short clip quote attributed to Jensen Huang referencing NVIDIA investors and a joke/aside that “that’s why Korea is rich.” No explicit product-cycle, earnings, guidance, supply-chain, or valuation claim. Actionability is low; at most it’s sentiment/brand-strength context for NVDA and a vague nod to Korea’s tech/investor base.
Very short clip quote attributed to Jensen Huang referencing NVIDIA investors and a joke/aside that “that’s why Korea is rich.” No explicit product-cycle, earnings, guidance, supply-chain, or valuation claim. Actionability is low; at most it’s sentiment/brand-strength context for NVDA and a vague nod to Korea’s tech/investor base.
About this channel
Short-form market commentator active on X as @jukan05. Writes on AI, semiconductors, tech capex, and related talent/supply-chain issues. Commentary ranges from quick product references (e.g., “Z Fold 7”) to thematic investment implications (AI datacenter capex, TSMC execution moat).
@jukan05
Most recognized assets
Unlock the full track record
Follow @jukan05 on X for concise, thematic takes on AI and semiconductors; use posts as idea-generation rather than direct trade instructions—many entries are narrative or rumor-level and not explicit trade calls.
8 more thesis calls are available after sign-up.