Reinforcement Learning is hard..

Caleb Writes Code· 125K subscribers

12d ago · English audio

FireCrawl: https://firecrawl.link/caleb-writes-code Why RL doesn't scale and is difficult to set up. Reinforcement Learning has been helping LLMs improve capabilities in post training beyond what we could expect from these systems. And it is growingly becoming more important in robotics, gaming and simulation but where are the bottlenecks and challenges that still exist? What about RL makes it so tricky and what do…

Watch on YouTube

ABOUT THE CHANNEL

Caleb Writes Code

Part editorial, part informational on AI. I'm passionate about inference, hardware and infrastructure, model architecture, economics, AI agents, and yes, coding. Videos will cover a wide variety of topics around AI and how to make sense of this changing world.

3.0×Score at discovery

173.2KViews

57.2KMedian · 50 videos

Outlier score

This score was measured on Tue, 03 Nov 2026 07:35:56 GMT. Current view counts may differ.

172.6K views ÷ a 57.2K baseline = 3.0×. Compared with 50 earlier videos from the same channel, matching the format recorded at scoring time. Older-upload baseline: reference videos had at least as long to gain views. This comparison is not age-matched.

YouTube snapshot: Tue, 03 Nov 2026 11:10:14 GMT. Counters may have changed.

Read the methodology

Artificial Intelligence & Data

Channel-balanced percentile: 93

Comparison details

153 videos from 91 other channels, with each channel weighted equally. Same format and language, with 100,000–999,999 subscribers. Observed at 296 hours after publication, within ±48 hours.

The comparison uses this broader topic because the specific niche has too few examples. This describes the collected sample, not all of YouTube. Updated Tue, 03 Nov 2026 12:00:41 GMT.

Niches & tags

Primary niche

Predictive models, model training, evaluation and learning algorithms.

Video tags

reinforcement learning RL training RL suck RL is hard why RL is hard why RL is not scalable RL scalability RL budget how to RL LLM what is RL RL explain explaining RL how does RL work RL limitations RL bottleneck

Transcripts and detailed analytics are available for recent outlier Shorts up to three minutes. Research tools

Reinforcement Learning is hard..

Caleb Writes Code

Outlier score

Artificial Intelligence & Data

Niches & tags

Video tags

Breakouts from other channels

I Spent 100 Hours Building This Video Game With Claude (It's INSANE)

You Need to Learn How to Use Uncensored AI

Claude + Google Vids = Full Stickman Animations on Autopilot