cd /news/machine-learning/moving-from-ml-quant-work-into-robot… · home › topics › machine-learning › article
[ARTICLE · art-174243] src=dev.to ↗ pub= topic=machine-learning verified=true sentiment=↑ positive

Moving from ML/quant work into robotics: a concrete first week, measured

A developer transitioning from ML and quant work into robotics published a measured week-one log of reproducing a Soft Actor-Critic (SAC) baseline on MuJoCo's HalfCheetah using CPU-only PyTorch, reporting that each physics step takes roughly 50 microseconds while a SAC update at batch size 256 costs about 6 ms, making gradient steps rather than simulation the bottleneck. The work, scripted with an AI coding agent, includes per-run eval.csv logs, a plot script and a generated comparison table, released MIT-licensed in the sac-mujoco-repro repository, with a $1 packaged version offered to collect first reviews.

recommendation

Profile step versus update cost before committing to a training budget: here physics steps ran ~50 microseconds against ~6 ms per SAC update at batch 256.

more in playbook: Fine-tuning and data
by read1 min views1 publishedNov 22, 2026

Part of a public week-one log of moving from ML/quant work into robotics: what was actually run, what it cost, and the honest next steps. The scripting was done with an AI coding agent; every number below was measured on one CPU-only Linux box (Python 3.12, MuJoCo 4.0.1, PyTorch 2.15 CPU wheels).

The robotics job market talks about VLA and foundation models, but the

interviews keep probing the layer underneath: state estimation, kinematics,

contact, and the discipline of measuring before claiming. MuJoCo is free, runs on

a laptop CPU, and every experiment below is checkable by a reader in under an

hour.

pip install --target + PYTHONPATH + CPU torch wheels. Cost: about an hour, most of it reading error messages. The recipe is written down so it costs you ten minutes.env.step() on HalfCheetah: ~50 microseconds. So the bottleneck is gradient steps, not physics: a SAC update at batch 256 costs about 6 ms. Multiply before you commit. Everything above ships with logs: eval.csv per run, a plot script, and a README

whose tables are generated. The claim is never "I know robotics"; the claim is

"here is a working baseline, here is what it cost, check it yourself".

Everything above is checkable: code, full training logs and the generated comparison table are MIT-licensed at sac-mujoco-repro. If you'd rather skip re-running the 1M-step training yourself, the packaged version — logs, learning curves, checkpoints, README — is on our store for $1. The price is there to collect first reviews, not to gate the free repo.

── more in #machine-learning 4 stories · sorted by recency
── more on @mujoco 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/moving-from-ml-quant…] indexed:0 read:1min 2026-11-22 · —