Robonaissance
Subscribe
Sign in
Home
Notes
🧠 Intelligence Foundations
🧭 Agentic Intelligence
🤖 Embodied Intelligence
🌍 World Models
Archive
About
Latest
Top
Discussions
The Journey of RL, Part 2: The Value Hypothesis
Q-Learning and Its Discontents. The compression that made an entire field possible, and the trap it concealed.
Jun 9
•
Hugo
8
5
May 2026
The Journey of RL, Part 1: Before the Equation
How machines learned what to optimize, beginning in 1898 with a stopwatch and the borrowed assumption that has shaped reinforcement learning ever since.
May 26
•
Hugo
11
3
12
Tokenomics, Part 3: The Agent Tax
Agent tasks consume orders of magnitude more tokens than chat. The inversion lands on application P&L. Gross margin compresses. Where the cost lives…
May 20
•
Hugo
7
3
Tokenomics, Part 2: The Great Inversion. From Training to Inference.
Training capex was the story for three years. Inference is the story now. The largest pool of AI profit is forming downstream. Where the margin is…
May 18
•
Hugo
10
7
Tokenomics, Part 1: The Token Is the New Kilowatt-Hour
Token prices have fallen by two orders of magnitude in three years, more on the cheapest routes. The Token Stack. Margin Geography. Where the value…
May 16
•
Hugo
18
2
11
The Humanoid Value Chain: Where Does the Money Actually Go?
The trillion-dollar stack. Six layers. Three profit pools. Where the next decade of capital actually accrues.
May 14
•
Hugo
9
7
Inside China’s Machine: Unitree
Filed for a $7 Billion IPO. The Price Destroyer. 5,500 Humanoids Shipped. Now Profitable. Waiting on a Brain.
May 11
•
Hugo
8
7
The Making of DeepSeek
Ten years of quant infrastructure built the foundation. R1 made the company global. V4 extended the pattern. The original refusals are starting to bend.
May 6
•
Hugo
6
4
The Making of OpenClaw
Thirteen years in Vienna shaped the instincts. Three years of burnout proved the cost. A weekend hack passed Linux in three months. The architecture…
May 5
•
Hugo
3
5
The Rise of Agents, Part 8: The Open Frontiers
Eight articles. Two walls down. Three frontiers open. The agents got better. The summit did not move.
May 4
•
Hugo
2
4
The Rise of Agents, Part 7: Agent Meets World
Software agents act on text. Physical agents act on a world that does not accept undo.
May 2
•
Hugo
1
4
April 2026
The Rise of Agents, Part 6: The Agent Ecosystem
The breakthrough was making one agent work. The frontier is making many of them cooperate.
Apr 30
•
Hugo
2
3
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
