cd/sources/allen-institute-ai2· home› sources› Allen Institute (AI2)
cat /sources/allen-institute-ai2.feed | wc -l → 91

Allen Institute (AI2)

articles 91 domain medium.com → page 1/5 feed RSS
16:51
2026-09-29
medium.com
ai-agents

How to constrain a coding agent using only files you own

A developer working on harness and charter engineering reports that giving a coding agent a single ticket with no charter produced a run that passed only 1 of 8 checks, while adding a four-lever chart…

07:52
2026-09-19
medium.com
large-language-models

Bonsai: A 27B reasoning model on a 16 GB M2 Mac, with Ferrox

A developer has released Bonsai, a 27-billion-parameter reasoning model designed to run on a 16 GB M2 Mac, alongside a companion tool called Ferrox. The project targets local inference of a large reas…

17:49
2026-08-20
medium.com
artificial-intelligence

Can AI become conscious?

The Economist's recent article on AI consciousness prompted an engineer to argue that current language models are not conscious and that the debate is a distraction. The engineer contends that these s…

02:13
2026-08-05
medium.com
artificial-intelligence

The Moral Compass of AI: Why Fairness Shapes Our Future

A developer argues that fairness must be a foundational principle in AI development, warning that biased algorithms could perpetuate societal inequalities. The piece calls for proactive ethical design…

12:13
2026-08-04
medium.com
artificial-intelligence

Scaling pain arrives earlier when AI writes the code

A developer reports that AI-generated code accelerates development but introduces scaling challenges earlier in the product lifecycle, as the speed of code creation outpaces traditional infrastructure…

17:42
2026-08-02
medium.com
artificial-intelligence

AI-Powered Container Standardization

Capital One Tech has introduced an AI-powered container standardization tool that automatically normalizes Dockerfiles and container configurations to enforce best practices and reduce security risks.…

09:08
2026-07-31
medium.com
artificial-intelligence

Why Enterprise AI Gets Hard After the First Agent

DolphinDB Inc. introduced DolphinX, an enterprise AI platform designed to handle complex data analysis beyond simple chatbot interactions. The platform integrates large language models with the compan…

08:58
2026-07-29
medium.com
artificial-intelligence

From a Research Report to a Validated Factor with AI

DolphinDB Inc. demonstrates how AI can streamline the process of converting a research report into a validated quantitative factor. The company's platform uses large language models to build, debug, a…

17:23
2026-07-27
medium.com
developer-tools

The middle path between a stock agent and a custom harness

A developer proposes a middle path between stock coding agents and custom harnesses, introducing pi, an open-source terminal coding agent that ships a general terminal-agent harness with replaceable c…

07:30
2026-07-26
medium.com
artificial-intelligence

Let Local LLMs Search the Web Without Burning 100k+ Tokens

A developer created TinySearch, a system that allows small local LLMs to search the web without consuming the entire context window, avoiding the high token cost of 100k+ tokens.…

23:09
2026-07-23
medium.com
ai-safety

When AI Attacks AI: Why Every System Needs a Guard

A developer warns that AI systems are increasingly vulnerable to adversarial attacks from other AI, arguing that every AI deployment needs a guard mechanism to prevent manipulation and ensure safety.…

page 1 / 5 next →