🚨 Attention, Solana aspirants. Q3 Applications are closing soon. If you've been waiting, this is your sign to apply. This quarter, we have a majorly revamped Builders Program designed around the latest trends shaping the ecosystem. Learn by building alongside fellow builders, transform ideas into standout capstone projects, and build PoW that speaks for itself. @solana @solana_devs @SolanaFndn

Open-weight models don't just give you more control. They give you more responsibility. Choosing the model is only the beginning. To achieve data sovereignty, you need control over the infrastructure, governance, security, and data boundaries around your AI. Read why that matters:

These guys show what it's like when you bring institutional experience to Solana. Come apply to @solanaturbine if you've been thinking about bringing your edge to Solana and supercharging it into something goated.

Q2 Builders Capstone Demo - 7 kevred, by Léo Marquand and Yves-Marie (@kevredlabs) A BYOK Solana RPC load balancer. Users add their own provider API keys in priority order, and the proxy routes every request to the highest-priority healthy provider, failing over automatically when one goes down. watch the full demo here 👇

Q2 Builders Capstone Demo - 6 Bastion, by @rit3sh_x A policy firewall for @solana. Bastion lets a wallet owner grant scoped, short-lived, revocable authority to an AI agent or bot, with every action enforced on-chain. Watch the full demo here 👇

Tomorrow, SIMD-0286 activates at Epoch 1009. Here is how this would impact validators and applications. Validators: - Heavier maximum work done per block. If you are running on lower-end hardware, you could likely see increased execution times as well as replay degradation. - Ledger growth rises only to the extent blocks actually fill, so day-one impact is limited. But provision against the higher ceiling rather than current averages is recommended. The same goes for RPC providers, indexers, and archival nodes. Applications: - More transactions land per block. High-intensity programs like order-book DEXs get more room to operate before blocks fill, though the per-transaction cap of 1.4M CUs is unchanged. - Apps with distributed activities across accounts benefit the most. - Apps on one hot writable account are still bottlenecked as that account still caps out at 12M CUs. if SIMD-0306 passes, the per-account cap would move from a static 12M to 40% of the block limit (40M at 100M), which would actually relieve hot-account congestion. - More block space means less fee competition during congestion, so lower priority fees and fewer retries. If your app hardcodes fees or estimates from 50M/60M data, revisit it, otherwise you'll systematically overpay. - Scheduled/automated execution such as agentic txns are going to be more reliable during congestion. For app developers: nothing breaks, no migration needed, existing transactions and blocks stay valid, and it is important to highlight that the throughput gain from this is contingent on validators keeping up.

Three weeks ago I shipped driftwatch and said there was a lot to improve. Since then it ran on my validator for 3 days. It caught the disk freezing for over a second, 110 times. The drive's own health check said everything was fine the whole time. here's everything I found🧵

Turbin3 builders was amazing. So glad to be part of this community Any companies or devs who are using transfer hooks in your workflow. Feedback on what you would want in a service like this will be amazing. Check at @jettysolana I have updated it a lot since this was presented.

How low can Solana's slot times realistically go? 100ms? 10ms? Lower? @MaxResnick from @anza_xyz took our builders deep into how @solana actually reaches consensus. → Simplex & the skip-vote → Gatling & arbitrarily small slot times → Multiple concurrent proposers (Braid, MinCP, Constellation) → Asynchronous execution → What Alpenglow changes for all of it and plenty more There's always something special about getting Max in a room with our builders, the questions go deep, the back-and-forth gets real, and everyone walks out sharper. This is the frontier of Solana consensus, explained by one of the brilliant minds building it. Watch the full session now🔗👇