News
Curated articles, my blog posts, and site announcements.
All Categories
Loading news...
Dria News
Company-specific announcements, X posts, and blog updates from the portfolio.
Portfolio news
43
Exo7 news items
Lighter13 news items
Lucis5 news items
MegaETH14 news items
Monad19 news items
Morpho16 news items
Prime Intellect39 news items
Accountable8 news itemsDria7 news items
Edison4 news items
Friend4 news items
Octet2 news items
Runner7 news items
Agora5 news items
Aligned Layer17 news items
Delta2 news items
Fabric Cryptography8 news items
Giza17 news items
Praxis5 news items
Rhinestone9 news items
Sorella10 news items
Succinct21 news items
Wildcat9 news items
Zenith7 news items
OWN6 news items
Phylax26 news items
Berachain12 news items
Clique10 news items
Herodotus8 news items
Inco25 news items
Intuition6 news items
Pimlico15 news items
Ritual34 news items
Confetti1 news items
Eclipse18 news items
Knox14 news items
Mizu6 news items
Mode9 news items
Movement13 news items
Nillion5 news items
OpenQ3 news items
PIN AI12 news items
Pragma16 news itemsSort:
- AI
Dria launches Watchmen, an open-source skill writer for coding agents
Dria released Watchmen, a local open-source tool that writes shared skill files for Claude Code, Codex, and pi from a user's own sessions.
- Research
Dria uses Goodfire interpretability tools to study developer code style
Dria published research using Goodfire AI's interpretability tools to test whether individual developer code style is detectable inside a model trained on that developer's code.
- AI
Dria launches Kai, an AI engineer for codebase maintenance
Dria unveiled Kai, a long-horizon AI maintainer that reviews PRs, benchmarks performance on real GPUs, and catches production regressions across GitHub, Modal, Vercel, Linear, and Jira.
- Research
Dria publishes long-horizon agent benchmark research
Dria compressed 30 days of long-horizon agent operation into a 12-minute benchmark and shared three findings on memory, tool drift, and feedback loops.
- Research
Dria publishes memory architecture research for persistent coding agents
Dria detailed how capping agent memory at 100 entries per category and prioritizing curation over accumulation outperformed naive memory growth in long-running coding agents.
- Research
Dria experiments with self-directed fine-tuning on Qwen3.5-4B
Dria pointed its KAI research agent at Qwen3.5-4B and let it author its own fine-tuning curriculum, demonstrating self-directed model training using Thinking Machines tooling.
- AI
Dria releases Mem Agent for stateful local memory across LLMs
Dria launched Mem Agent, a 4B model with local interoperable memory across Claude, ChatGPT, and LM Studio that turns stateless LLMs into stateful agents with persistent human-readable memory.