The problem of large LLM memory footprint doesn't seem to have a scalable architectural solution. Full attention is a necessity for modern LLMs. I ran a genetic algorithm over LLM architectures hyperparams. They all lose on retrieval vs plain attention. The exception I wanted isn't there. jasonnormore.com/blog/search…

turbopuffer crossed $100M run-rate in March. 19mo after $1M. Profitable & <$1M raised. Cursor・Anthropic・Notion・Cognition・Harvey・Bridgewater・Ramp・Linear・Legora・Superhuman・Atlassian・Granola We’d be nowhere without them. We work like hell to exceed their expectations.

Been on a quest to make models run with a smaller memory footprint, especially locally on my laptop, so I built a hybrid SSM attention model with three memory layers: • minimal attention KV for turns • SSM state, MB-scale memory footprint for episodic • LoRA from captures for long-term learning Long context without the memory-bandwidth tax of GB KV caches, no prompt inflation. I'm curious to see how it scales to larger models. github.com/jnormore/cognit

Back with another product update courtesy of my favourite @JonnieSimpson. Manually triggered flows are now live on @heymantle LFG. 🚀 Until now, every flow in Mantle ran on a schedule: daily, weekly, on a trigger event. Useful, but not every flow wants to be a cron job.

New on @heymantle: Marketing Sites 🔍 Mantle's AI already helps you write docs and accelerate customer support for your app. But the full customer journey? It was a blackbox until now. We shipped tracking pixel for Marketing Sites.

A while back we shipped one-click affiliate payouts at @heymantle. Everyone loved it. 💛 But we kept hearing the same thing: "I still have to remember to log in and do it." Turns out, even one click is one too many. So we took the last click away.

New on @heymantle: Smarter email suppression 🎯 We just shipped a small feature with a big impact on deliverability. You can now automatically suppress role-based email addresses like support@, info@, noreply@ from your marketing sends.

You don’t always need a bigger LLM, just more diverse ones. So I built an ensemble inference proxy that sends prompts to multiple small models in parallel and combines their responses. Initial results look great! gpt-4.1-mini haiku qwen 3b (local): 74% accuracy. GPT-5 alone: 73%. Claude Sonnet: 74%. This ensemble config is 13x cheaper and 2.5x faster than GPT-5. And I haven’t even tested other providers yet. The trick: cross-provider diversity. Same-family ensembles do nothing. But models from different providers make different mistakes, and that's exploitable. Tested 27 configurations across 6 aggregation strategies. The best ensemble beats GPT-5 on knowledge tasks by 8 percentage points. Easy to experiment with your own configurations, just a YAML and emerge sweep. github.com/jnormore/emerge

I did a quick run with this on shopify/liquid @tobi - 10% reduction in allocations and GC time in about 15 minutes with 27 iterations. Will try it with a longer run at some point but very promising first try. It proposed: - Replace markup.scan regex with byte-level iteration - Added caching for VariableLookup objects (biggest impact) - Added .freeze to all byteslice returns - Replaced markup.strip with byte-level whitespacing skipping ⁃ Inline the create_variable method call