Robotics
Buying robots: the questions that matter
For anyone evaluating an automation proposal, the specification sheet is the least useful document in the pack.
Tobias Nkemelu
22 articles by Nkemelu
Robotics
Safety standards and how robots are allowed near people
The rules governing industrial robots are specific, well developed, and the reason most machines still work behind a fence.
AI & Compute
What to actually worry about with AI
A great deal of the risk discussion is about scenarios, and a shorter list of problems is already causing measurable harm.
Robotics
Legged robots: what they are for
Quadrupeds became a commercial product before humanoids, and the applications are narrower and more sensible than the videos suggest.
AI & Compute
Where AI systems actually fail in production
Not in the model. In the data pipeline, the distribution shift, the feedback loop and the assumption that the world stays still.
AI & Compute
Evaluating an AI product claim
A short checklist for reading announcements, which mostly consists of asking what was measured and against what.
Robotics
Agricultural robotics and the labour that is disappearing
Fields are unstructured, produce is delicate, and the economics are being forced by a labour shortage rather than pulled by capability.
AI & Compute
Small models, distillation and the case against always using the biggest
Capability per parameter has improved faster than parameter counts have grown, which changes the economics of what to deploy.
AI & Compute
Retrieval, context and grounding a model in your data
Supplying source material at inference time is the most reliable fix for fabrication, and it moves the difficulty into the retrieval step.
Robotics
Surgical robots and what teleoperation actually buys
The most commercially successful medical robot is not autonomous at all — it is an extremely good remote control, and that turns out to be the point.
AI & Compute
Agents, tools and the gap between demo and deployment
Giving a model tools and a loop makes it far more useful and introduces reliability problems that compound with every step.
AI & Compute
Open weights, closed models, and what the distinction means
Releasing model weights is not the same as open source, and the practical consequences differ from the licensing argument.
Robotics
Industrial robots: what the installed base actually does
Several million arms are working in factories today, doing a narrower set of tasks than most people imagine.
AI & Compute
Training data, copyright and the unresolved question
Models are trained on text and images their creators did not license, several jurisdictions disagree about whether that matters, and the courts are still working.
AI & Compute
How models are actually deployed
Serving a model in production is a systems problem — batching, caching, routing and cost — and it looks nothing like running one on a laptop.
Robotics
Warehouse robotics: the automation that actually happened
While humanoids get the attention, wheeled robots quietly became one of the largest deployed automation categories in the world.
AI & Compute
What a GPU actually does, and why it is scarce
Matrix multiplication in parallel, memory bandwidth as the real bottleneck, and a supply chain with a single point of concentration.
AI & Compute
Benchmarks and what they fail to measure
A score on a public test set tells you about the test set. Contamination, saturation and construct validity all get in the way.
Space
Satellite internet: what it does and does not solve
Low-orbit constellations genuinely changed rural connectivity, and the physics sets limits that marketing tends to skip.
Robotics
Humanoid robots and the demo-to-product gap
A robot that walks across a stage is solving a different problem from a robot that works an eight-hour shift.
AI & Compute
Why AI needs so much electricity
Training is a large one-off cost. Inference is a small cost repeated an enormous number of times, and it is the one that scales with success.
AI & Compute
What a large language model is actually doing
Next-token prediction at scale, which explains both the capability and the failure modes better than any metaphor about thinking.