Tech sommelier & AI Whisperer Always a GameDev 🎮 @ ❤️ Mostly blind🦯, He/Him, All opinions my own
England, United Kingdom
Joined June 2013
- Tweets7.8K
- Following3.6K
- Followers1.5K
- Likes25.6K
Pinned Tweet
Time to open source a few of the libs I've been working on.
Very much built with AI rather than by AI.
Inspired by the cool stuff in @QuestDb & @TigerBeetleDB
This is bolt, the data piece left over from my lock down game engine experiments
github.com/Ugbot/bolt
And yes it's in #cpp 😜
One of my favourite activities is replaying the broken sword games with my children.
We just finished number 3.
Though I do now want to remake it in 2d based on its concept art....
Go to gadget claude!
Finally building the Asset managment and editors now :D
asset import pipelines with AI assisted collision shapes are proving their weight in gold. Still all in C with a dash of lua for UI stuff
Considering that that the local GPUs cost a non trivial amount of power, maintainence, and the models update slower it's crazy
For the cost of 8 years of $200 Claude subs, you can force your team to use shitty quantized versions of DeepSeek v4 Flash running at 30tps. Incredible deal!
🐿️Ben Gamble🦯 retweeted
Join the Startup Talent Q&A panel organised by @AntonKooll - where @TalentAscent will be sharing practical advice and tips on talent strategy for #ScaleUps
🗓️ 12 noon BST - 12th October
📍Virtual (MS Teams)
🔗 Sign up via @LumaHQ luma.com/esz39rkn
🐿️Ben Gamble🦯 retweeted
Meet Mistral Large 4, aka Le Chonk.
• 1T parameters, natively multimodal. 49B active.
It is the best open weights model from US or Europe on aggregated benchmarks.
• State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding.
• Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure.
• Available to all via API today. Working with cybersecurity partners privately.
Open weights release end of October.
🤖 Made with AI
Rendering is one of those things where the complexity doesn't so much spiral out of control as you have to base jump into caves full of obscure math, vendor controlled APIs and performance traps.
After some fixes, and a scattering update, my renderer clocks in at about 113k lines of code. 20k of which could be cleaned out by removing openGL3.3 and 4.6 leftovers.
Turns out that this is actually quite small, @o3dengine's atom renderer clocks in at about >380k lines. abstractions between metal, vulkan & DX11/12 being a very large part of it. atom being actually quite new and modern. I haven't run the tools over @UnrealEngine yet, Any guesses on LoC?
Building procedural generation systems with Vulkan and C.
Taking some time to build out some of the stuff I've had on the shelf for a while. With AI handling the build system, its a lot more fun :D
🐿️Ben Gamble🦯 retweeted
Three years ago, we started Restate because we felt that building reliable distributed applications was just too hard. We were solving the same problems around failures, recovery, and consistency again and again.
Today, I'm excited to share that we've raised a $20M Series A led by Singular with participation from Redpoint Ventures and Capital One Ventures.
The need for solutions like Durable Execution has only grown since we started. Today, the majority of new Restate use cases are AI agents: long-running processes, coordinating across many tools and APIs, that need to keep working correctly through all sorts of failures and interruptions, coordinate sub-agents, or schedule work.
At the same time, our customers have pushed Restate into places that expanded our own view of what Durable Execution could become. Replit moved the execution of Replit Agent onto Restate. DOSS runs high volumes of enterprise workflows on Restate and Fortune 500 institutions are using Restate for complex transactional workflows.
We think this is just the beginning and excited to build what we believe will be a fundamental pillar of all infrastructure in the future.
Read more here: restate.dev/blog/announcing-…
🐿️Ben Gamble🦯 retweeted
Agents are developers that need their experience to be dialed-in and seamless, just like a human developer.
Which means you still need DevRel, or a form of it, to win in the agentic ecosystem, too.
🐿️Ben Gamble🦯 retweeted
We believe that the future is code + AI, so made workflow evals to reflect that
Jev costs: $42 / BILLION input tokens ($0.042 / MTok) and output tokens are free (forever - they’re too cheap to meter with our new architecture)
Jev is named after Jevons paradox and off the intelligence per $ charts
