Bookmarks

Bookmarks

49953 bookmarks
Custom sorting
Mellea — Reliable, Testable LLM Output for Python
Mellea — Reliable, Testable LLM Output for Python
Open-source Python library for structured, validated LLM output. Type-annotated outputs, token-level constrained decoding, requirements-driven validation. Works with any LLM provider.
·mellea.ai·
Mellea — Reliable, Testable LLM Output for Python
How we scale PgBouncer in ClickHouse Managed Postgres | ClickHouse
How we scale PgBouncer in ClickHouse Managed Postgres | ClickHouse
PgBouncer is single-threaded, so a single process caps out at one CPU core no matter the box size. See how ClickHouse Managed Postgres runs a peered fleet of PgBouncer processes with so_reuseport to scale pooling across every core — with benchmarks showin
·clickhouse.com·
How we scale PgBouncer in ClickHouse Managed Postgres | ClickHouse
In defense of polyfills • Lea Verou
In defense of polyfills • Lea Verou
A highly influential spec editor expressed the opinion that polyfilling is harmful. I beg to differ.
·lea.verou.me·
In defense of polyfills • Lea Verou
Do we still need build tools?
Do we still need build tools?
Do we still need Rspack, SWC, PostCSS, Vite, Parcel, Webpack, Turbopack, esbuild, Rollup, Rolldown, Babel, Autoprefixer, esbuild and Lightning CSS in 2026?
·olliewilliams.xyz·
Do we still need build tools?
Fetch Needs Error Codes
Fetch Needs Error Codes
HTTP/2 and HTTP/3 have rich, typed error signaling that the Fetch API completely discards. The TC39 error code proposal could fix that.
·jasnell.me·
Fetch Needs Error Codes
Telegram Serverless
Telegram Serverless
Telegram Serverless lets you run backend code for your bot and Mini App directly on Telegram's infrastructure — no servers…
·core.telegram.org·
Telegram Serverless
tinbase — the Supabase-compatible backend that fits in a tin
tinbase — the Supabase-compatible backend that fits in a tin
Local Supabase dev without Docker — one process, real Postgres with Row Level Security, Auth, Storage, and Realtime, and it even runs in the browser. The official supabase-js SDK works unchanged.
·tinbase.dev·
tinbase — the Supabase-compatible backend that fits in a tin
Introducing Open Interpreter
Introducing Open Interpreter
A new open-source coding agent for low-cost and open-weight models.
·openinterpreter.com·
Introducing Open Interpreter
soycaporal/ternlight
soycaporal/ternlight
Contribute to soycaporal/ternlight development by creating an account on GitHub.
·github.com·
soycaporal/ternlight
The Pulse: What can we learn from Bun’s rapid Rust rewrite with AI?
The Pulse: What can we learn from Bun’s rapid Rust rewrite with AI?
To a sceptic, spending $165K to migrate Bun from Zig to Rust sounds very expensive. But to a realist, shortening a 1-2 year migration down to 11 days opens amazing new opportunities for devs.
·newsletter.pragmaticengineer.com·
The Pulse: What can we learn from Bun’s rapid Rust rewrite with AI?
Ant, a lightweight JavaScript runtime
Ant, a lightweight JavaScript runtime
Run real npm packages and TypeScript - all from a single 9 MB binary. Ant is a lightweight JavaScript runtime built from scratch on its own engine, with a Node-compatible API, a VM-isolated sandbox, and Wasm support built-in.
·antjs.org·
Ant, a lightweight JavaScript runtime
HTMX and Web Components Instead of React
HTMX and Web Components Instead of React
Veteran CTO who built successful companies through empathic leadership and mastery of software delivery - from sustainable code to stakeholder trust. Now exploring the future of empathic and sustainable tech.
·kore-nordmann.de·
HTMX and Web Components Instead of React
Sebastian Raschka on X: "For agentic coding, one can say: - Unless you need Terra Ultra perf, it's always better to use a Luna model with higher effort setting (same or better performance but cheaper). - Forget everything below Sol High, use Luna with higher effort settings here - Forget Sol Extra https://t.co/Tjc9mELCer" / Twitter
Sebastian Raschka on X: "For agentic coding, one can say: - Unless you need Terra Ultra perf, it's always better to use a Luna model with higher effort setting (same or better performance but cheaper). - Forget everything below Sol High, use Luna with higher effort settings here - Forget Sol Extra https://t.co/Tjc9mELCer" / Twitter
- Unless you need Terra Ultra perf, it's always better to use a Luna model with higher effort setting (same or better performance but cheaper). - Forget everything below Sol High, use Luna with higher effort settings here - Forget Sol Extra
·x.com·
Sebastian Raschka on X: "For agentic coding, one can say: - Unless you need Terra Ultra perf, it's always better to use a Luna model with higher effort setting (same or better performance but cheaper). - Forget everything below Sol High, use Luna with higher effort settings here - Forget Sol Extra https://t.co/Tjc9mELCer" / Twitter
JUMPERZ on X: "so this is what i've found works best with gpt-5.6 so far.. > luna high, normal everyday coding, fast, capable, doesn't feel wasteful... >luna xhigh better quality without jumping to the expensive models... >terra medium, bigger features >terra high, repo-wide changes.. > https://t.co/CvBee4zCqq" / Twitter
JUMPERZ on X: "so this is what i've found works best with gpt-5.6 so far.. > luna high, normal everyday coding, fast, capable, doesn't feel wasteful... >luna xhigh better quality without jumping to the expensive models... >terra medium, bigger features >terra high, repo-wide changes.. > https://t.co/CvBee4zCqq" / Twitter
> luna high, normal everyday coding, fast, capable, doesn't feel wasteful... >luna xhigh better quality without jumping to the expensive models... >terra medium, bigger features >terra high, repo-wide changes.. >
·x.com·
JUMPERZ on X: "so this is what i've found works best with gpt-5.6 so far.. > luna high, normal everyday coding, fast, capable, doesn't feel wasteful... >luna xhigh better quality without jumping to the expensive models... >terra medium, bigger features >terra high, repo-wide changes.. > https://t.co/CvBee4zCqq" / Twitter
Context engineering with Dex Horthy
Context engineering with Dex Horthy
Dex Horthy explains why context engineering is key to building more effective AI-assisted software without sacrificing code quality.
·newsletter.pragmaticengineer.com·
Context engineering with Dex Horthy
All you need is PostgreSQL
All you need is PostgreSQL
Introduction The setup Laying the foundation The foundation: schemas and user roles for modularity Domains Accounts, managed and external Transfers, constrained by a state machine and temporal periods Transfer state history Account auditing Transactions, the immutable events On maintaining business rules via meaningful constraints The transfer state machine Transactions must fall within the transfer period Pending transactions require a pending transfer No future transactions when closing a transfer On capacity planning Working set estimation On write throughput Enabling HOT Updates for Transfers Making sure there are no Unused indexes OLTP Listing The history of a transfer OLAP Balance ledger Incremental maintenance via triggers On serializable isolation On decoupling Benchmarking the startup scenario Seed data Write script: full transfer lifecycle Read script: activity stream and balance Running the benchmark Results Conclusion Appendix A: Full code suite. Introduction There is a deep cultural reflex in modern engineering: whenever a problem appears, reach for a packaged solution instead of thinking from first principles. The result is architectural cargo culting and lots of missed opportunities. Some intentionally absurd-but-familiar examples:
·ebellani.github.io·
All you need is PostgreSQL