Long-form analysis at the intersection of AI and life sciences — all spells, sorted by time.
Auto-Quant lets LLMs autonomously iterate on FreqTrade strategies via backtesting loops. Interesting autoresearch concept, but deeply locked into crypto trading infrastructure. Only useful if you live in the FreqTrade ecosystem.
Fifty years of antioxidant trials mostly failed because 85% of mitochondrial DNA damage comes from replication errors, not free radicals. From mitochondrial transplantation and MOTS-c peptide hormones to CoQ10's irreplaceable role in the electron transport chain, aging science is moving from 'slow the damage' to 'replace the parts.
agentic-markdown embeds variable substitution and conditional blocks into Markdown syntax. A neat concept, but with only 15 GitHub stars, no loop support, and mature alternatives like Jinja2 already covering the same ground, it is hard to recommend for any serious workflow.
An assistant professor shares how an AI-powered literature management pipeline transformed manual PubMed searching into an automated digestion workflow, systematically processing 50+ academic papers per week through tiered reading.
Auto-PPT turns each slide into a React TSX file that AI agents can read and edit directly, no GUI or screenshot loops needed. Solid concept and well-designed agent protocol, but still early-stage (29 stars, 0 releases). Worth tracking, not yet worth depending on.
Starting from the everyday experience of watching ChatGPT type out words one by one, this article uses six everyday analogies to break down KV Cache, Paged Attention, Flash Attention, GQA, Continuous Batching, and Speculative Decoding — explaining how large language models serve more users faster on the same GPU.