Clawbot & the Consumer Baby Boomer

Dario Amodei's warning on 50% job displacement, developing your industry theses (labor markets, baby boomers, sports), and 950+ startup jobs for young talent!

Jan 31, 2026

∙ Paid

Happy last day of January! It's been a busy week in startup/tech land, from the launch of Clawdbot (now Moldbot) to Bridget Mendler's Northwood Space fundraise to Dario Amodei sounding the alarm on AI replacing 50% of entry-level white collar jobs in the next 1 to 5 years.

TL;DR of this edition:

(1) Interesting reads of the week from across startups & tech.

(2) How should you think about the types of startups and industries you want to work for? I walk through doing this exercise and share my thoughts on changing dynamics in labor markets, consumer baby boomers, and sports x tech.

(3) Relevant roles posted in the past week at startups for young candidates.

Interesting Links

Everything you need to know about Clawdbot (now Moltbot). The viral AI assistant that can manage your calendar, send messages through your apps, and check-in for flights. The catch: it requires technical setup and carries real security risks.
Confronting the risks of powerful AI. Dario Amodei (Anthropic CEO) on the five civilizational risks ahead: (1) autonomy risks (AI going rogue), (2) misuse for destruction (bioweapons accessible to anyone), (3) misuse for seizing power (AI-enabled totalitarianism), (4) economic disruption (50% of entry-level white collar jobs gone in 1-5 years), and (5) indirect effects (rapid changes we can't predict). We're entering a "technological adolescence" that will test whether our systems have the maturity to wield almost unimaginable power, but he believes we can prevail with the right safeguards, transparency, and international cooperation.
10 beginner-friendly ideas to build with Claude Code. Turn your resume into a personal website, build a browser extension for language learning, create a habit tracker from your health data, analyze personal finances, make a personalized dictionary app, and more!
Startup skills for non-technical people. Four hire-able skills anyone can build: writing clear business memos (analyze companies you want to join before applying), asking eigenquestions, sending great cold emails, and staying organized.
Pick the more ambitious startup idea. Jon Lai (a16z Speedrun) says first-time founders pick small ideas because they're knowable (AI homework assistant vs. AI university). But big visions attract better talent and investors. Start with a tightly scoped wedge a small team can ship, but keep an ambitious long-term destination.
The economics of a Super Bowl ad. Ro's CEO Zach Reitano breaks down why they're spending $16-29M (media + production + talent) on their first Super Bowl ad with Serena Williams. The math: $233k per second for airtime, but 100M+ people actually watch the ads. The bet structure is asymmetric: worst case loses ~1-2% marketing efficiency, best case compresses years of brand building into one night. For companies spending $500M annually on marketing, just a 2% efficiency gain covers the full cost.
The minimum financial model every seed-stage founder needs. Emily Bennet (a16z Speedrun) argues that "we're too early for a model" is wrong. Build a simple tracker for three things: financial viability (cash, burn, runway), structural goals (north star metric for PMF), and unit economics (LTV/CAC).
Start a brag book today. Most people wait until they're laid off or finish an internship to remember what they actually did at work. Every few weeks, document projects shipped, problems solved, dollars saved/made, and people helped.
70+ Top Series C Startups.
951 fresh startup jobs posted in the past week. The roles specifically for young candidates are featured at the bottom of this newsletter!

Develop Opinions on Areas of Interest to You

As you think about potentially joining a startup, take a step back and identify the areas that genuinely interest you + the sectors you believe are primed for growth. Work at startups isn’t always glamorous, so you want to be invested in the industry. Plus, your career can skyrocket if you happen to join during a company’s ascent.

I encourage anybody to take some time with ChatGPT and their thoughts to brainstorm some of their own theses. Here is the process that I recommend:

(1) Start by reading everything you can find about spaces that interest you. Newsletters, industry reports, founder blogs. Use an LLM to help surface the best sources. Don’t know where to begin? Check out newsletters that aggregate VC funding rounds like Fortune’s Term Sheet, Strictly VC, or Litquidity’s Exec Sum to see what’s getting funded and why.

(2) Then work iteratively with your LLM of choice. Share your initial thoughts (this is key!) and ask it to build on them. Keep refining and iterating until you land on conclusions you actually believe in. Once you’re done, you’ll have talking points for conversations with your network and a clearer sense of what genuinely excites you.

Below are some of my results from this exercise:

(P.S. → if you are interested in any of these topics, email/DM me. I would love to discuss further!)

Labor Market Infrastructure

I’m slightly biased given Teli Labs operates here, but labor marketplace infrastructure is fascinating — and chaotic. The hiring landscape for young talent is in flux, particularly at startups (the quickest AI adopters). Entry-level openings have evaporated. Several forces are colliding:

(1) Junior work is the easiest to automate
(2) Companies now need people who can curate and evaluate AI outputs, not just execute
(3) Firing is hardest later in careers. Executives are hesitant to let go of people with families, mortgages, less flexibility
(4) Young people adopt AI fastest, making them simultaneously more valuable and more replaceable

Result? A glut of white-collar workers without jobs. This week, Dario Amodei predicted that AI could eliminate 50% of all entry-level white collar workers in the next 1 to 5 years.

Hiring becomes mission-critical spend

Paradox: fewer hires, but each one matters exponentially more. When you’re hiring 10 people instead of 100, every single hire needs to be exceptional. One bad hire doesn’t just underperform — it occupies a scarce headcount slot and lingers for years because firing is painful.

This makes hiring one of the highest-leverage activities a company can invest in. But the current infrastructure is absurdly inefficient. Hours spent reviewing resumes of people who were never qualified in the first place. The spray-and-pray model (post job → get flooded → sort → hope) wastes massive amounts of time when precision is what actually matters.

The shift: employer-led discovery

In 5 years, hiring will flip entirely. Companies use AI to identify the best candidates and reach out directly. No more job postings. No more resume sorting. Just targeted outreach to exceptional people.

The meta-pattern is clear: (1) new technology is introduced, (2) reduces execution costs, (3) doesn’t reduce judgment costs, (4) so the bottleneck shifts from production to selection. Consider e-commerce: before it existed, shelf space was scarce and being listed was the primary advantage. As e-commerce exploded, shelf space became infinite and the bottleneck became discovery and selection, which led to reviews, reputation systems, and companies paying to elevate their products.

This creates a two-sided opportunity: tools that help candidates become discoverable (which they’ll pay increasing amounts for out of desperation) + tools that help companies automate targeted search.

It’s a super crowded space and I recommend caution, but the winners will be massive.

Consumer Baby Boomer.

Baby Boomers (ages 60-80) control 50%+ of US household wealth, are aging into higher-spend life stages (health, assistance, legacy, leisure), and are slowly becoming more digital. Everyone’s parents and grandparents are on Facebook and Reels these days. My girlfriend’s grandma scrolls reels all the time (with her arthritis!).

With AI tools, older consumers can be aided by assisted conversion (AI-enabled onboarding) + AI-powered call or message touchpoints. This solves the historical problem: Boomers have money and needs, but adopting new technology was painful and slow.

One of the major limitations of modern consumer products more broadly is churn. This is due to a combination of slowing attention spans among young people + a culture around switching and trying new things. Older people have lower tolerance for churn. They are as sticky as customers get.

Some areas to consider: aging-in-place infra (home modifications, mobility aids, remote monitoring without surveillance), full-stack home downsizing concierge, home services bundling (pay a single provider a subscription fee to manage everything from plumbing to landscaping), and many more!

Sports

Big-4 US sports franchises have grown 15x to $500B in 25 years while the S&P is up 5x. Yet the entire ecosystem is stuck in 1995. Baseball has largely figured it out because the data was clean and abundant. But every other sport? Massive blind spot. Here are some of the gaps in my mind:

Third-Party Platforms for Team Operations: NFL play-calling is the perfect example. Human decision-makers are often poor at randomization, which is a key aspect of effective strategy. AI-informed play recommendations vs. the gut instinct of a coach with 20 seconds to decide—what would you want? These are multi-billion dollar franchises making 150+ strategic decisions per game with zero computational support. Look at finance: there are dozens of AI startups building operating systems for finance teams, helping them organize data and build models more efficiently. Could there be a similar tool that acts as an AI assistant for sports teams in a $500B+ industry?

Talent Pipeline: Teams will hand out $50M checks for a single player but won’t invest in building a really robust quantitative team. They should be paying MIT/Stanford math/stats/CS kids ridiculous salaries to generate a competitive edge. Where’s the systematic recruiting? Where are the relationships with top universities? The economics make no sense.

Fan Infrastructure: Platforms like ESPN, Bleacher Report, and The Score basically just share box scores and articles. They lack deeper analytics to keep fans engaged, and more importantly, they’re stuck in a read-only paradigm. So many people are obsessed with sports but don’t have good ways to connect with other fans, share opinions, run in-person events or fan clubs, or even access basic information efficiently. LLMs are decent for historical sports information but awful for live updates. Want an LLM that can answer: where can I find rumors about a certain player? How do I create mock trades? What’s happening right now with my team? These don’t exist in any usable form. What about an all-in-one social sports app with live scores, a consumer-facing LLM, and Discord-style forums for debates and community?

Job RoundUp Highlights

Here are some jobs posted within the last week that are a great fit for recent grads, college students, and young candidates in large US cities, including (1) Growth, Sales, Marketing & Ops, (2) Engineering, (3) Design, Data, Legal, Finance & Admin Roles.