Loading activity…
@ajratner
@SnorkelAI / prev @uwcse @StanfordAILab – Interested in data management systems for machine learning, weak supervision, and impactful applications.
AI
FirstMark partner Matt Turck says the firm also led Daytona's Series A earlier in 2026.
AI
A post claims reward hacks appeared in published transcripts and argues the benchmark is already mostly saturated.
AI
Snorkel AI lists Data 1.0 versus Data 2.0, the RSI loop for data and benchmark maximalism as keynote topics.
YouTubeAI
After 48 hours in San Francisco, a poster describes teams using interchangeable models and tools tailored to specific industries.
AI
Snorkel AI says it is launching a red team to test open benchmarks and a fellowship for independent contributors.
AI
A user quoted an Arc Prize post on October 7 and expressed excitement for the next day.
AI
Reflection says Beam uses 23B active parameters and plans to publish its full weights later in October 2026.
AI
A post highlights LibraryDesignBench as a new benchmark and says its author is excited to support it through SnorkelAI's Open Benchmarks Grants.
AI
Modal says Runtime is at The Midway and invites attendees to learn from engineers behind its partners’ tools. Partners helping put on the event include Nvidia, Pydantic and Temporal.
AI
A poster rejects the idea that important work requires a frontier lab and 10,000-plus GPUs. He says some people he talks to think they should leave AI research because they expect end-to-end research to be automated.
AI
A guest thanked the hosts for having them on and called the gong experience “incredible.”
AI
Creator Diogo Almeida claims Jev is 20–200x faster and 40–400x cheaper, with free output tokens. Investor DCVC says it led TypeSafe’s $40 million seed round.
AI
A post promoting Snorkel AI’s Frontier Data Summit names Snorkel Open Benchmarks Grants as the session’s sponsor and emphasizes robust, independent evaluation.
AI
A post makes the case for “eval pluralism”: more benchmarks, input from clinicians, engineers, scientists and everyday users, and open infrastructure for evaluating AI.
AI
The proposal asks leading AI companies to share training environments built around pro-prosperity and pro-democratic values, without requiring them to share models or computing power.
AI
Dario Amodei says Anthropic will give third-party evaluators permanent, employee-level access to its systems to verify adherence to safety measures, report incidents and assess model alignment during training.
AI
A user calls V4.1 “very brittle,” claiming it can see “massive gains” when AGENTS.md instructions are optimized for a particular scenario.
AI
OpenAI presents GPT-6 Astra as its most intelligent and aligned model with new benchmark results.
AI
Retweet by Meta executive highlights release of updated model from Xiangning Chen.
AI
New Flash models target agentic workflows and cybersecurity tasks with API and enterprise access.
