My Morning LLM Mech Suit

We are so back

The AI bulls are back! Insiders believe AI is here to take your dev job, but the hiring market suggests otherwise… * Austin Allred Tweet [https://x.com/austen/status/1927880370646589830?s=12&t=WxI040bG2OGvl1V7bEHARg] * Software Engineering Jobs Tax Deduction [https://x.com/ghosttyped/status/1927420463455609261] * Anthropic researchers “white collar jobs automated in 5 years” [https://x.com/ns123abc/status/1925939773279019322] * Google Chief AI Scientist - “AI = Junior dev in less than a year” [https://x.com/unusual_whales/status/1923876433975476486] * Lenny’s Job report [https://www.lennysnewsletter.com/p/state-of-the-product-job-market-in]

My Morning LLM Mech Suit

Cheat on everything and your degree means... nothing? Don't give into the extreme viewpoints, everything will be just fine. * Exploring technological disruptions, from cheating tools to AI's impact on education and software development. * Cluely's launch sparks ethical and legal debates about cheating in academic and professional settings. * 49% of Gen Z feel college degrees are devalued as employers prioritize AI literacy over traditional qualifications. * LLMs boost coding speed in familiar frameworks but require human oversight for quality in complex projects. * Only 20% of seasoned Big Tech engineers embrace AI tools, with 30% considering leaving the industry. * How can we balance AI-driven progress with integrity and fairness?

College Admissions

Our discussion moves to college admissions, where many surprising rejections have been recently coming through. Also, a new week, a new model with Llama 4 and people are seriously questioning AI progress. * Llama 4 [https://ai.meta.com/blog/llama-4-multimodal-intelligence/] * AI Progress Questioned [https://www.lesswrong.com/posts/4mvphwx5pdsZLMmpY/recent-ai-model-progress-feels-mostly-like-bullshit] * Analyzing the Housing Affordability Crisis for Engineers [https://www.reddit.com/r/Salary/comments/1jojlwa/engineers_dont_make_good_money_anymore_part_2/?utm_source=share&utm_medium=mweb3x&utm_name=mweb3xcss&utm_term=1&utm_content=share_button] * The Overwhelming Complexity of Software Engineering [https://0x1.pt/2025/04/06/the-insanity-of-being-a-software-engineer/] * Cal AI founder rejections [https://e.vnexpress.net/news/news/education/18-year-old-millionaire-ceo-of-ai-company-rejected-by-6-ivy-league-schools-4870460.html] * And corresponding tweet [https://x.com/zach_yadegari/status/1906859987105636667] * High-Achieving Student Rejected by Ivy League for CS, Accepted to UMass Amherst [https://x.com/deedydas/status/1906898551273337209] * College Admissions Competition [https://x.com/felpix_/status/1905502809916596376?s=12]

Copium

In this one we chat about recent AI developments, including Gemini 2.5 and ChatGPT's image generation, along with a deep dive into one person's experience on the levels.fyi forum. * Gemini 2.5 pro is cooking? [https://simonwillison.net/2025/Mar/25/gemini/] * ChatGPT image generation [https://simonwillison.net/2025/Mar/25/gemini/] * Blocklists in the industry [https://www.businessinsider.com/how-block-lists-affect-your-job-search-and-career-advancement-2025-3] * Is the market just cooked for new grads? [https://www.levels.fyi/community/thread/9qYUTN/is-the-market-just-cooked-for-new-grads] (from levels.fyi [https://www.levels.fyi/community])

Extreme Vibe Coding

Oops, we failed recording episode 6. This one worked! Vibe coding is the new hotness. Is it really the future? We talk about (more) CEOs declaring software careers dead early and why the AI takeover is not as clear cut as some would have you think. * Student using AI to cheat on interviews [https://www-cnbc-com.cdn.ampproject.org/c/s/www.cnbc.com/amp/2025/03/09/google-ai-interview-coder-cheat.html] * WSJ says job hopping isn’t as lucrative as it once was [https://www.wsj.com/lifestyle/careers/salary-deflation-workers-new-jobs-0b2e1a1f] * Anthropic's CEO predicts AI will code 90% of software in 3-6 months, all in a year [https://www.businessinsider.com/anthropic-ceo-ai-90-percent-code-3-to-6-months-2025-3] * Bear case against AI taking over [https://www.lesswrong.com/posts/oKAFFvaouKKEhbBPm/a-bear-case-my-predictions-regarding-ai-progress] * Sesame new voice assistant [https://www.sesame.com/research/crossing_the_uncanny_valley_of_voice#demo] * https://www.ultravox.ai/ [https://www.ultravox.ai/] * Karpathy coining the term "Vibe Coding" [https://x.com/karpathy/status/1886192184808149383]