Keyword matches words; semantic matches meaning and exact titles or names; hybrid combines both. In keyword mode, use quotation marks for exact phrases. Author matches published credits. Year applies to posts; wiki entries have no author or publication year. You can also use author:"Ada" or type:podcast in non-regex searches.
author:"Ada"
type:podcast
Explore articles, conversations, and the wiki.
Articles, conversations, profiles, and grounded writeups.
Select Apply filters to update results.
Search or filters have changed. The results below still show the previous search; select Search or Apply filters to update them.
1341 results
…3 Opus, best of 4 runs. We are working on clustering and flow engineering with Haiku. New AI Models and Benchmarks Contextual Position Encoding (CoPE): @jaseweston introduced CoPE, a new positional encoding method for transformers that…
…those addresses with readable page content could give the model context and precise control. One participant argues that reasoning is already sufficient and that representation and control are the main remaining engineering problems.…
…insights into technical mechanisms behind naming, prompt engineering, or AI behavior in self-identification context. No statistics, code, or external references provided. It's gotten to the point where I notice chatGPT's linguistic style…
…smooth progress: apparent walls and plateaus may give way to engineering advances, with several areas of improvement contributing at once. For builders, the answer distinguishes two kinds of work: - Compensating for model limitations: some …
…Brex’s team now consists of ~300 engineers and ~350 across EPD (Engineering, Product, & Design). Their team is split across product domains in addition to infrastructure. Their core product domains are the following: Corporate Card…
…about human users. An agent with much larger memory and context might benefit from different interface choices. Query design supplies a concrete case. One speaker previously preferred building a REST endpoint for a known need. Agents’ less …
…strengths in coding, instruction following, and handling long contexts (up to 1 million tokens). @kevinweil notes that GPT-4.1 achieves a 54 score on SWE-bench verified. Instruction Following: @OpenAIDevs points out that GPT-4.1 follows…
…for malware injection or other harmful behavior. The engineering question is where to enforce limits on the agent’s actions; the heuristic itself is no guarantee of safety. Brev faced a more physical visibility problem at GTC: late…
…messages🔥🔥): Hallucination vs. Real Progress in GPT, Prompt Engineering vs. Session Engineering, Context Window Limits and Memory, External Databases for Context, Importance of verifying Facts with GPT *GPT Progress Reports: Hallucination…
…and where a particular vendor’s product fits. Forward deployed engineering addresses both by placing vendor engineers in customer environments to advise on useful applications and co-develop a solution. The guest views that substantial…
…ecosystem. Even where the rhetoric is heated, the substantive engineering question underneath is real: how much of current risk is solvable with control, oversight, sandboxing, and org process versus requiring slower capability…
…His thesis went beyond detecting isolated objects or events. Context lets detections reinforce one another: seeing sky above land can increase confidence in both classifications. Limited labels and a small corpus constrained this contextual…
Sign in with Google, or get the newsletter by email.
Loading Google sign-in…
A conversation with Greg Brockman about OpenAI’s path to AGI, alongside our analysis of GPT-5, Codex, and the shift to agentic coding.
Thariq Shihipar on Claude Code’s next era: prompts, harnesses, plugins, and agents—and what it takes to keep pace with the frontier.
Marc Andreessen on the future of the browser, Pi and OpenClaw, and why this wave of AI feels different.
Priscilla Chan and Mark Zuckerberg on Biohub’s plans for frontier AI and virtual biology, and the ambition to cure or prevent disease.
Anima Anandkumar on modeling the physical world, from weather to fusion reactors, and why physics needs more than language models.
John Platt on automating science, tackling climate change, and how the next generation can contribute in the age of AI.
Anthropic brings interactive apps into Claude with an open MCP standard for richer AI interfaces.
What Sam Altman’s town hall reveals about building startups in the age of AI.
Moonshot’s multimodal model brings competitive coding, native image and video understanding, and parallel agents.
Free. Change your preferences or unsubscribe any time.
Uncheck all to create an account without newsletter emails.
Discover technical newsletters we recommend, then explore Latent Space membership. Both are optional.
Your newsletter choice is optional. How we use your account data.
Please enable JavaScript to sign in or subscribe. You can still read the articles.