Sources · Twitter/X

@ankrgyl

Active

ankrgyl

Linked items 125
Errors 20
Last fetched 6h ago
Priority P3
Must-read No

Edit

Human-friendly label. Leave blank to fall back to the identifier.

Drives ingestion cadence and the default ranking weight applied to items from this source.

Includes every eligible item from this source in the must-read section, even when the normal item cap or duplicate grouping would skip it.

Automatic import processes at most 50 historical items per burst. Full history continues in resumable chunks and may incur transcription/LLM cost. Older items are never silently dropped.

Progress: 0 processed · 0 remaining · 0 skipped · status none

Optional day window for classification drain when using Latest 50.

Cancel

Recent items

The 15 most-recently-ingested items from this source, newest first.

Title Type Fetched
@ankrgyl: A related learning is that LLMs tend to patronize themselves. So if you ask them to write skills, or... article 2026-08-08
@ankrgyl: there's a lot of leverage left on the table by llm-generating skills instead of doing that, try han... article 2026-08-07
@kyliestew: got to sit down & talk about how we use evals to make sure our dashboard agent actually solves r... article 2026-08-07
@daRubberDuckiee: I ran an eval comparing Kimi K3 vs Claude Sonnet 5 as the agent model for frontend design tasks. Wh... article 2026-08-07
@braintrust: Cloudflare's dashboard agent spans its entire developer platform, from deploying Workers to debuggin... article 2026-08-07
@mattturck: How to build long-horizon AI agents: behavior specs, ontologies, process supervision - my conversati... article 2026-08-06
@ankrgyl: Congrats to @howietl and @airtable. Building a product that is a joy to use backed by a business wit... article 2026-08-04
@ankrgyl: Wrote up some thoughts on the differences between evals and tests. They're not the same, and evals ... article 2026-08-04
@ankrgyl: Article Evals ≠ tests While building Braintrust I’ve been thinking about what evals are good at,... article 2026-08-04
@ankrgyl: one of the best feelings when you're building a complex product is to solve the harder version of a ... article 2026-08-04
@ankrgyl: software engineering right now is all about getting out of the agent's way, securely article 2026-08-03
@ankrgyl: early observation with deepseek v4 flash it's super smart but i have to encourage it to keep workin... article 2026-08-02
@ankrgyl: everyone should follow @mitch_troy. there are very few people who are actually building agents (vs. ... article 2026-08-01
@ankrgyl: one of the ironies of commercial open source software is that it's usually harder to self-host than ... article 2026-08-01
@ankrgyl: im starting to think we're conflating "i need a sandbox service" with "i need a sandbox" the proble... article 2026-08-01