Search Latent Space

Results for “context engineering”

Articles, conversations, profiles, and grounded writeups.

?
Search tips

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.

Reset filters

Select Apply filters to update results.

1341 results

not much happened today

…map functionality reiterated its dependency on relevant code context, crucial for accurate code modifications tagged to identifiers. As per Paul's clarification, the model evaluates identifiers based on their definitions and references,…

AI News · 2024-10-24
AI News—2024-10-24
Hamel Husain

…assistant for real-estate professionals. Early prompt engineering produced rapid progress, but expanding the assistant’s responsibilities exposed a plateau: fixing one failure introduced others, prompts accumulated edge cases, and the team …

Person · 2026-09-20
Person—2026-09-20
GPT-5 Codex launch and OpenAI's quiet rise in Agentic Coding

…(53 messages🔥): Workflow use case variation, Prompt engineering using steps and semi-programming language, ElevenLabs conversation agents handle the system prompt, Breaking GPT5, Dynamic context Model choice varies by workflow use case:…

AI News · 2025-09-15
AI News—2025-09-15
Heather Kulik

Heather J. Kulik is MIT’s Lammot du Pont Professor of Chemical Engineering, with a joint appointment in Chemistry. Her research combines quantum chemistry and machine learning to guide molecular and materials discovery. In a 2025…

Person · 2026-09-17
Person—2026-09-17
Claude Opus 5: Fable-level performance at Opus price (half Fable)

…and skills (Activity: 971): Anthropic’s post, “The new rules of context engineering for Claude 5 generation models”, says they cut over 80% of Claude Code’s system prompt for newer Claude models with no measurable coding-eval regression,…

AI News · 2026-07-25
AI News—2026-07-25
High Taste Testers: Making AI Assistants Useful in Practice

…Another called remote execution and its interface a hard engineering problem. A third pointed to a possible ingredient for moving a session between machines: externalize its state in an append-only log, so another environment can resume…

Writeup · Date unavailable
Writeup——
ChatGPT Advanced Voice Mode

…output could potentially improve the format. Clarifying prompt engineering: Discussion involved defining prompt engineering, highlighting the importance of clearly stating requests and providing enough context. One member noted that prompts…

AI News · 2024-09-25
AI News—2024-09-25
How AI Moats Accumulate: Glean, Anthropic, and OpenRouter

…acknowledging that many are only a couple of years old. A lab’s engineering and distribution may make entry possible; whether entry is strategically wise remains a separate decision. The career discussion describes academic competition in…

Writeup · Date unavailable
Writeup——
AI Engineer World's Fair: Second Run, Twice The Fun

…Friday :) The first World's Fair was a big experiment - is AI Engineering big enough to warrant its own large multitrack conference? We were fortunate to be the first to preview capabilities that now every single AI Engineer can build with …

AI News · 2025-05-07
AI News—2025-05-07
How Computer-Use Agents See, Act and Recover

…try again? At OpenAI DevDay, Weinstein, who leads product and engineering for computer-use agents, and API product lead Nikunj Handa describe two connected pieces of the stack. Weinstein discusses how an agent operates software; Handa…

Writeup · Date unavailable
Writeup——
GPT 5.4: SOTA Knowledge Work -and- Coding -and- CUA Model, OpenAI is so very back

…SOTA for tool/GUI operation (OpenAIDevs; sama). Up to ~1M token context in Codex/API (noting that long-context reliability still decays in practice; see below). Efficiency / “fewer tokens, faster speed” framing (OpenAI), plus later addition…

AI News · 2026-03-06
AI News—2026-03-06
How NotebookLM Was Made

…sources of 500,000 words each, too much to fit into the model’s context window at once. Handling that collection across text and images remained an unresolved engineering problem; no detailed retrieval design was presented. Natural audio…

Writeup · Date unavailable
Writeup——
Your place in Latent Space

Join Latent Space.

Sign in with Google, or get the newsletter by email.

Loading Google sign-in…

or subscribe by email

Free. Change your preferences or unsubscribe any time.