The Adil Wali Show

AI agents are starting to take over the work people find pointless. Davide Ghia on why a lot of that work is how juniors learn the job, and who will be senior enough to supervise the agents.

1. 0:00 From individual contributor to manager of AI agents
2. 3:14 "Bullshit jobs," measured task by task
3. 5:03 A bullshit job, or just a bad job?
4. 8:33 Meaning is a perception: status, energy and people
5. 12:07 If AI can do everything, what should it take first?
6. 15:03 Why the boring tasks may be how you recharge
7. 19:54 The four-box: what workers want to hand off vs. what AI does well
8. 21:52 Cognitive endurance: when AI takes your downtime
9. 24:17 Why the two-day work week is a myth
10. 25:04 From doing to supervising: who trains tomorrow's experts?
11. 28:48 Where will the next generation of leaders come from?
12. 34:06 Small teams gain; big-company teams are at risk
13. 38:55 Will competition keep people employed?
14. 43:33 Can AI make apprenticeship better?
15. 48:47 AI sounds confident. Who's willing to say no?
16. 57:53 Mental atrophy: staying sharp when you only supervise
17. 1:03:25 Does anyone ask workers what they want to automate?
18. 1:05:56 Doing right by workers as a business strategy

Timed transcript · Full show notes and sources

• Will AI Agents Free Us From Meaningless Work? A Human-Centered Analysis (Ghia, Ranjit, Cerquitelli and Quercia) — arXiv

Adil Wali invests in startups. At 20:02 he describes how he looks for companies to invest in. At 21:12 and 31:09 he talks about OpenAI and Anthropic. Nothing in this episode is investment advice.

The grid doesn’t have an energy problem; it has a peak-power problem. Chris Dunlap on why AI data centers face a choice between connecting now and flexing in the worst hours, or demanding firm power and waiting years—and which AI work can actually wait.

1. 0:00 Is AI already paced by power?
2. 2:48 From F-18 backseater to the power grid
3. 4:15 Why is China out-building the US on power?
4. 6:10 How a Northern Illinois power crunch became the paper
5. 7:34 Why can’t the US connect power plants faster?
6. 9:56 What PJM is and why the queue is jammed
7. 12:50 Lawyers vs engineers: why regulators exist
8. 14:12 Energy vs power: why the peak matters
9. 15:45 If supply can’t catch up, can demand flex?
10. 19:16 A peak-power problem, not an energy problem
11. 21:47 Why flexible data centers still can’t jump the line
12. 23:25 Which AI work can wait? Urgency, surge and spot pricing
13. 28:35 Moving AI to wherever the power is
14. 32:04 What would make workload shifting possible?
15. 36:11 Who has to play ball: states, moratoriums, and your power bill
16. 41:22 Utilities, RTOs, and paying data centers to flex
17. 45:21 What to watch: will hyperscalers trade flexibility for speed?
18. 50:18 Can the US build fast enough to compete?

Timed transcript · Full show notes and sources

• Quantifying AI data center flexibility as a resource adequacy asset — Research Square (preprint)
• Quantifying AI data center flexibility as a resource adequacy asset (working paper) — The Institute for Climate and Sustainable Growth, University of Chicago

Adil Wali invests in startups. At 50:53 he mentions that startups he has invested in are moving workloads to Chinese open-weight models. Nothing in this episode is investment advice.

Open source you share. Proprietary you buy. Maxim Khailo argues the next era is software you can’t buy at all—and states the one condition that would end it. Plus why a deployed model doesn’t learn, and the energy bill under every agent fleet.

1. 0:00 Why I wanted to talk to Maxim
2. 1:14 From HERE Maps and a quant fund to the limits of deep learning
3. 11:18 What has actually changed since 2012 — and what hasn’t
4. 15:13 How we got here, and why a deployed model doesn’t learn
5. 25:25 100 million agents would draw the entire energy of the UK
6. 26:13 Where does the harness end and the model begin?
7. 31:50 Thetix, and what efficient context actually buys you
8. 40:20 Why the CLI became the operating system for agents
9. 51:37 “We’re entering the era of private software”
10. 57:41 The caveat — this only holds while the tokens keep flowing
11. 1:00:02 The energy crunch underneath all of it
12. 1:05:52 Is there a moat in the models?
13. 1:13:17 Closed-loop learning, and the hardware it will take
14. 1:28:24 Agent fleets, and where orchestration breaks down
15. 1:43:01 The Seattle boom, and what Maxim is most worried about losing

Timed transcript · Full show notes and sources

Adil Wali is an investor in Thetix. Maxim Khailo is associated with the company, and it is discussed in the episode from 31:49.

AI can make one engineer feel 100x faster. Jared Green explains why companies aren’t becoming 100x more effective—and why the bottleneck moves into review, coordination, technical debt, and the systems around abundant code.

1. 0:00 When code becomes abundant, what stays scarce?
2. 1:36 Why enterprise AI transformation stalls
3. 5:37 How to measure AI productivity
4. 9:29 The human pace car
5. 12:36 Jared’s unconventional path into engineering
6. 26:53 Microsoft, Windows Phone, and choosing the hard problem
7. 39:33 What AI changes about talent and teams
8. 45:51 The new operating model for AI engineering
9. 49:56 AI slop, engineering waste, and technical debt
10. 54:12 Are humans the bottleneck?
11. 59:10 Why Jared started Autar
12. 1:03:29 Stop shifting work. Eliminate it.
13. 1:13:16 Does AI absorb the next constraint?
14. 1:17:36 Do junior engineers still have a future?
15. 1:26:16 Leaving Big Tech to become a founder
16. 1:40:49 Where LLMs may hit their limits

Timed transcript · Full show notes and sources

Adil Wali is an investor in Autar. Jared Green is Autar’s founder, and the company is discussed in the episode.

The market erased roughly $272 billion from ten durable SaaS companies while assigning more than $1 trillion to frontier AI labs. I explain why that trade may be underestimating systems of record, authority, and operational moats.

1. 0:00 Most people are getting the AI trade wrong
2. 2:36 Rebuilding software is not replacing it
3. 3:49 Retention is evidence—not the conclusion
4. 4:51 The capital inversion
5. 6:46 Models have context; systems hold durable state
6. 8:32 CRM—the durable record and disappearing interface
7. 11:10 Three possible winners in CRM
8. 15:20 Installed does not mean economically central
9. 15:47 ERP and the authority moat
10. 17:25 Marketing—abundant creation versus the learning loop
11. 18:54 The strongest counterargument
12. 20:06 Keep the system of record model-neutral
13. 21:44 Training costs and distillation
14. 23:53 Diminishing returns on intelligence
15. 24:49 The operator test—state, permission, and feedback
16. 25:30 Predictions and what would prove me wrong
17. 27:22 The investment implications
18. 29:23 The question for software in the AI era

Timed transcript · Full show notes and sources

• Salesforce FY2026 Form 10-K — U.S. Securities and Exchange Commission: Salesforce reported trailing attrition of approximately 8%; the metric excludes Slack self-service and current-year acquisitions.
• ServiceNow 2025 Form 10-K — U.S. Securities and Exchange Commission: ServiceNow reported a 98% ACV-based renewal rate. Its definition can count ordinary downsells as renewed.
• Workday FY2026 Form 10-K — U.S. Securities and Exchange Commission: Workday reported approximately 97% gross revenue retention, excluding expansion.
• HubSpot 2025 Form 10-K — U.S. Securities and Exchange Commission: HubSpot reported approximately 88% gross customer-dollar retention and changed its treatment of partner commissions in 2025.
• Workiva 2025 Form 10-K — U.S. Securities and Exchange Commission: Workiva reported 97.2% gross revenue retention and 112.8% net revenue retention for 2025.
• New sales agents in Microsoft 365 Copilot — Microsoft: A concrete example of a seller working through an agent while Dynamics 365 or Salesforce remains the underlying record.
• HubSpot agents move to outcome pricing — HubSpot: Customer Agent and Prospecting Agent are priced by resolved conversation or recommended lead rather than only by access.
• Salesforce FY2026 fourth-quarter results — Salesforce: Early Agentforce bookings were primarily expansion within Salesforce's installed base.
• Workday FY2026 investor presentation — Workday: Scale and embeddedness evidence for the ERP and HCM discussion.
• Claude Cowork research preview — Simon Willison: January 12, 2026 is the base date used for the market comparison. The episode says “since,” not “because of.”

This episode reflects my personal analysis and investment views. I own and may buy or sell positions in public and private technology companies discussed here. The market comparison uses different pricing mechanisms and valuation dates for public companies and private AI labs, so it is directional rather than a synchronized index. This is not investment advice.

I rebuilt the product of a startup I invested in, then measured exactly what 471 million processed tokens, 64 prompts, and a week of judgment actually bought.

1. 0:00 471 million tokens
2. 0:20 Why I invested in Thetix
3. 2:26 The product today — and where the job breaks
4. 5:00 The product experience I rebuilt
5. 8:20 How it was actually built
6. 9:11 The receipt — tokens, time, and prompts
7. 10:28 The shape of the work
8. 13:36 What still has to be proven

Timed transcript · Full show notes and sources

• The S&P 500 and Trump Accounts — S&P Dow Jones Indices: S&P 500 total return figures through December 31, 2025.
• Nasdaq-100 Total Return Index fact sheet — Nasdaq: Ten-year return through June 30, 2026; a different window from the S&P figure.
• World Wealth Report 2026 — Capgemini: U.S. high-net-worth population as defined by investable assets.
• Survey of Consumer Finances — Federal Reserve: Direct ownership of publicly traded stock among U.S. families.

I’m an investor in Thetix. That’s part of why I care enough to go this deep. The redesign shown is an independent exploration, not an announced Thetix feature. Nothing here is investment advice; every portfolio example is product analysis.