02
Theme · 02 ★ Refreshed 24 May · NVDA validated

The Physical AI stack.

The mechanical & motion picks-and-shovels behind every humanoid, robotaxi and autonomous platform. 38 listed names mapped to the six layers it actually runs on — actuators, magnets, sensors, FPGAs, integrators and platforms.

You cannot 3D-print a strain-wave gear. You cannot software your way past an NdFeB magnet. Every humanoid in the field today carries 24-40 actuators, 60+ sensors, ~3 kg of rare-earth magnetics and a sensor-fusion FPGA that nobody else makes. Unlike the GPU race, the bottlenecks here have decade-long lead times.

Today's reads

★ Live signal: Robotics readout Book RSI 60.85, breadth 84.2%, 17 names overbought and 0 oversold. Prose generation requires ANTHROPIC_API_KEY — see SKILL.md. Thesis intact until data says otherwise.

Theme stats
As at 24 May 06:00 Dubai
Names tracked
38
Tier A
0
Tier B
38
Tier C
0
Live signals
6
Layers
6
Conviction mix across 38 names
Auto-derived from Robotics MASTER tiers · 24 May
Tier A · 0
Tier B · 38
Tier C · 0
Unranked · 0
Tier A — buyable now / structural Tier B — high conviction, add on pullback Tier C — coverage / watch
★ Tier B · 6 names

The highest-conviction robotics names.

Names where the bottleneck is physical and the supplier set is narrow. Strain-wave gears, RV cycloidal reducers, BLDC magnets, sensor-fusion FPGAs, BOM-named partners — all with hyperscaler-grade demand visibility and 24-36 month lead times. Click any card for the full thesis.

● Active buy signals · 6 names

Fundamentals and technicals aligned.

Names where research conviction meets a live technical entry — scanner ENTRY signal, or trending above 50MA with RSI 40-65. These are actionable now.

★ Jensen's May 20 Q1 FY27 call · validated

Our robotics thesis survived the earnings test — and was conservative.

We mapped the 38-name Physical AI basket on May 14. Six days later Jensen ran the same playbook on the NVIDIA earnings call. The framework was right — what surprised us was the velocity. We modelled this as a 2027-28 P&L story; the print shows it is already a $9B TTM line two years ahead of the curve.

Physical AI TTM rev
>$9B
first hard number disclosed
Edge Computing Q1
$6.4B
+29% YoY · new segment
Uber robotaxi
2028
~30 cities · 4 continents
Software stack
6 products
GR00T · Isaac · Cosmos · Thor · Alpamayo · NuRec
02 · 06The Stack

From motor to manipulation.

Six physical layers, one narrowing aperture. Each layer has its own physics, its own moat, its own catalyst calendar — and a different bottleneck dynamic.

ALL 38 L1 6 L2 8 L3 6 L4 10 L5 1 L6 7
01
Layer · 01 / 06

Platforms / OEMs.

6 names mapped to this layer of the Physical AI stack.

02
Layer · 02 / 06 · ★ the bottleneck

Actuators & Motion ★.

8 names mapped to this layer of the Physical AI stack.

03
Layer · 03 / 06

Compute / FPGA.

6 names mapped to this layer of the Physical AI stack.

04
Layer · 04 / 06

Sensors / Vision.

10 names mapped to this layer of the Physical AI stack.

05
Layer · 05 / 06

Materials & Power.

1 names mapped to this layer of the Physical AI stack.

06
Layer · 06 / 06

System Integrators.

7 names mapped to this layer of the Physical AI stack.

03 · 06Today's Brief

Build · Trim · Coiled.

Three rotating buckets from the latest scan + research overlay. Each card carries entry · stop · invalidation and a thesis grounded in the robotics-stack physics.

Build · Entries4
454910 Robotics Build
454910
277810 Robotics Build
277810
ALGM Robotics Build
ALGM
RR Robotics Build
RR
Trim · Take Profits1
9880 Robotics Trim
9880
Coiled · Setups1
SYM Robotics Coiled
SYM