Sergio is live and building out and taking the iii platform and harness from zero to fully built out and deployable products with DeepSeek v4.1 Flash!
https://t.co/mu4x2skDWf
lol I just remembered when GitHub announced they were storing code in the Arctic. Feels like decades ago, back when code was actually something worth preserving
Asked iii to orchestrate the work with deepseek and delegating GUI Researcher and Front-end engineer to Astra. I asked it to create a Balatro looking Tic Tac Toe and it did!
@EdszSeren0@luker2o@AkitaOnRails É isso que eu penso também, as gigantes so estão se importando em colocar mais dinheiro em infra, poucas estão pensando em otimizar a própria tecnologia para utilizar menos recursos
@luccacerf O foda é que animação sempre precisa de muitas e isso pelo que percebi é fácil de automatizar. Inclusive quero testar o Astra fazendo uma cutscene no Unity
@luccacerf Sim, peço para usar o Blender MCP, mas percebo que o Astra cria muitas automações em Python. Ainda estou em testes muito iniciais, mas até agora ta muito promissor, consegui gerar animações muito boas só com uma foto do concept designer
People are trying to build their own harnesses, iii can give you the bear bones for anyone to build a harness that suits each specific need in every area
Build and own your harness, folks.
Very few people understand the magic behind customizing and optimizing an agent harness for your work.
You can start seeing incredible results quickly, even with a minimal harness. Just better code, better outputs, better costs, better writing, and so on.
As an example, I see way too many people complain that they are tired of their out-of-the-box agent harness producing verbose outputs. If you built a harness, that's easier to fix with a system prompt optimized for you.
This is why Pi is getting insane adoption these days. You want to control your harness. I agree that it's hard to optimize where things stand, but self-improving algorithms will make it easier to learn from trajectories. And if you use local models, you can tune all of that end-to-end (through a co-evolution of the harness and the models) as you work with them.
Look into open-source projects like Eve, Exo, and Prime Agent, and you will see what I mean.
@goon_nguyen https://t.co/kXDVST10sg
I think the problem might be the harness, this was 5h session with lots of prompts and I built quite a few stuff, all live. It costed $2 with over 1,000 API requests
Sergio is live and building out and taking the iii platform and harness from zero to fully built out and deployable products with DeepSeek v4.1 Flash!
https://t.co/mu4x2skDWf
@sophiamyang https://t.co/kXDVST10sg
Watch this one, we built iii and this will be the next big thing for AI. It's a long session, I live streamed yesterday, but will post a shorter summarized video probably today
Sergio is live and building out and taking the iii platform and harness from zero to fully built out and deployable products with DeepSeek v4.1 Flash!
https://t.co/mu4x2skDWf
Sergio is live and building out and taking the iii platform and harness from zero to fully built out and deployable products with DeepSeek v4.1 Flash!
https://t.co/mu4x2skDWf