Trending on X about Biohacking
Discover what's trending on X in longevity, health, optimization
This Week's Trending Activity
Explore what's trending on X about Biohacking. These are the most engaging Biohacking discussions trending on X right now. Our AI analyzes what's trending to surface the best content, updated weekly.
What's Trending on X about Biohacking
CATHIE WOOD CALLS OPTIMUS HUMAN-LEVEL BY 2028 ON 1GW COMPUTE SURGE ARK Invest's 2026 robotics deep dive just laid out a bold timeline: Tesla Optimus reaching human-level task proficiency around 2028, riding a sustained AI compute explosion to 1.08 gigawatts....

𝐈 𝐝𝐞𝐬𝐢𝐠𝐧𝐞𝐝 𝐚 𝐟𝐢𝐭𝐧𝐞𝐬𝐬 𝐚𝐩𝐩 𝐡𝐨𝐦𝐞 𝐬𝐜𝐫𝐞𝐞𝐧 𝐭𝐡𝐚𝐭 𝐚𝐧𝐬𝐰𝐞𝐫𝐬 𝐨𝐧𝐞 𝐪𝐮𝐞𝐬𝐭𝐢𝐨𝐧 𝐢𝐧𝐬𝐭𝐚𝐧𝐭𝐥𝐲: “Am I active today?” Clear stats, strong visua...

“Content ideas are not the bottleneck.” We spoke to a regenerative systems developer - permaculture/ living landscapes. Active professionals are not lacking ‘what to say.’...
Aging isn’t what slows men down. Hormonal decline is. After 30, testosterone trends down year after year unless you support it....
Aging isn’t the enemy. Neglect is. Testosterone drops year after year if you don’t support it....
Your body doesn’t stop making testosterone with age. It becomes less efficient at it. Shilajit helps improve the efficiency of hormone production by supporting cellular energy and enzyme function....
velocity is all that matters. speed. we have to go fast. faster than we think is possible. the individual is too slow. the swarm is the only organism fast enough to survive what's coming.

Swiss Entrepreneur Has Found Yet Another Use For AI – "Suicide Pods" That Determine if a Patient is Mentally Stable Enough to Die s.barstool.link/c/article-3561…

Prediction: the energy and compute hardware (chips, cooling, etc.) bottleneck will incentivize the evolution of models that are increasingly more energy and compute efficient. I suspect we’ll cross the barrier of (mostly) self-improving AI before we unlock major upscales of energy and compute, which means at least three things become likely: 1.) Self-improving AI will begin to increase energy-efficient and hardware-efficient intelligence rapidly, bringing massive unlocks in edge devices like AR glasses, live video AI, and robotics way faster than with only human development and way before energy and hardware start to scale exponentially. Likely in 2026-2028....
This can only be done if the inference is tracked and attributed. And only if the model usage is tracked and attributed. And only if the training data is tracked and attributed....
finally beat opus 4.5 on anthropic take home challenge, it is truly kernel optimization fun

Your body repairs itself when it can relax. Magnesium supports deeper sleep, where repair actually happens. Poor recovery accelerates aging faster than stress....
Spending 60k over five years “biohacking” only to realize taking 10x the govts recommended dose of vitamin D cures all your problems
You don’t stop producing testosterone with age. It becomes 1% less efficient every year after 30. Shilajit helps improve the efficiency of hormone production by supporting cellular energy and enzyme function....
This paper shows LLM training can stay stable at batch size 1, if Adam is set right. They tune Adam by token count, so batch size 1 can train an LLM stably. That cuts memory use and tuning work, and it makes full fine-tuning more realistic on 1 GPU....

H₂ Water for Diabetes: Transform Ordinary Water in Minutes into a Diabetes-Fighting Superdrink—the Scientifically Backed Home Solution to Treat the Cause, Not Just Symptoms
Anthropic killed prompt engineering courses. They have an open source guide (linked below) that's the holy grail for using Claude. Once you get past the technical blockers, the real skill issue will be clarity (knowing what you want). Spend 10 mins daily training that muscle.

Building agents continuous to evolve. As models grow smarter, we grant more autonomy & infrastructure. 2023/2024: Just workflows, chaining of LLM calls with inputs/outputs. 2025: Tool Calling, selection of 10-15 domain tools, 10-20 turns. 2026: Sandbox env with guardrails and 5-10 atomic tools. First signs of continuous learning...
I'm 6 years into building. I thought rest was for people who weren't serious. I pushed through fatigue for 3 years....
A well known important feature to stabilize RL training is implementing the LM head in fp32 precision to help with gradients. Reproduced the plot from the MiniMax M1 paper entirely on my dgx spark and in Ai2's post-training research codebase.

A new analysis using millions of job ads argues that the popular story that generative AI is already wiping out junior white-collar jobs does not match what hiring data shows. FT published a report. Hiring for both junior and senior roles slowed in mid 2022, months before ChatGPT launched....

Low IGF-1 doesn't magically mean you'll increase your lifespan. Yes, it drives: - Tissue repair...
If you were smart or lucky enough to have not been poisoned by the US Government's deadly COVID "vaccine" (death jab, that causes cancer, heart attacks and strokes) you should be able to buy life insurance at a much reduced rate. Hopefully we can end rigged elections and we can...
Youtube used to show you a ton of aggregate stats on your watch history, but removed it a while back. I built a Youtube Rewind (link in next tweet) that you can run to see lifetime stats including: - Videos Watched - Top Channels...


























