teslajoy.bsky.social

@teslajoy.bsky.social

https://teslajoy.github.io/bio/ I use AI as augmentation for resolving complex problems, not alienation of human connection.

Neat Perspective. I think it captures something we bump into often. Instead of, "Can AI learn biology?" Break it down: • What biological structures are learnable at each level? • How do those levels relate? • How do they give rise to biological states and function? Worth a read.

Why machines don’t speak biology: Toward native biological language models

Advances in AI have shown promise in protein structural information, but in this perspective, the authors argue that biology’s true complexity—emergent, evolved, and irreducibly messy—demands more tha...

cell.com

Really like the idea behind Epistemic Blinding: arxiv.org/abs/2604.06013 Language in question matters so much! + LangSmith evaluators over blinded/unblinded trace pairs. + Lightweight post-hoc audits for long-running agents. + A reusable audit protocol anywhere entity names carry strong priors.

Epistemic Blinding: An Inference-Time Protocol for Auditing Prior Contamination in LLM-Assisted Analysis

This paper presents epistemic blinding in the context of an agentic system that uses large language models to reason across multiple biological datasets for drug target prioritization. During developm...

arxiv.org

🔥 topic at AI Dev Conf 2026: complexity + durability + evals the shift isn’t just agent design - it’s runtime systems durability, state, observability, eval loops agents that run > agents that just demo

The Runtime Behind Production Deep Agents

Deploying long horizon agents in production requires purpose-built infrastructure. This guide covers durable execution, memory, HITL, observability, and how deepagents deploy ships it all to productio...

langchain.com

High performers who are easy to work with can sometimes be quietly pigeonholed. This happens more often than we admit. Reliability should open doors, not define the ceiling. Choose environments where your happiness is tied to growth, not containment. Onward.

Just one data engineer in a vast community nudging the science snail forward 🐌💻➡️💨 .... building backend bioinformatics platforms and services to solve real problems, not reduce life to “data experiences.” The Extinction of Experience by Christine Rosen? So True. Respect. wwnorton.com/books/978039...

The Extinction of Experience

An <em>Esquire</em> Best Book of 2024 <br /><br /> A reflective, original invitation to recover and cultivate the human experiences that have atrophied in our virtual world., The Extinction of Exper...

wwnorton.com