Michael Short

tech guru, father, & entrepreneur. always working on cool stuff. prev - @microsoft & @paloaltontwks

[Michael Short @mcjaytown]

Optimized throughput and latency via:

- Cross-node EP-powered batch scaling
- Computation-communication overlap
- Load balancing

Working hard or hardly working 🤔🇺🇸

[Michael Short @mcjaytown]

@believeapp pms

[Michael Short @mcjaytown]

@dextorocoin @believeapp @dextorocoin @DexToroApp is launched by serial scammers who have launched three separate dextoro tokens on their twitter. check previous CAs.

https://t.co/o8x53Np8It

5weKScrfJaLEjpP7UctqqYgZDEvVRFHyK7gjtadJpump

[Michael Short @mcjaytown]

@KRinweb3 @AppStore @phantom @dextorocoin @DexToroApp is launched by serial scammers who have launched three separate dextoro tokens on their twitter. check previous CAs.

https://t.co/o8x53Np8It

5weKScrfJaLEjpP7UctqqYgZDEvVRFHyK7gjtadJpump

[Michael Short @mcjaytown]

@DexToroApp @dextorocoin @believeapp @pasternak @dextorocoin @DexToroApp is launched by serial scammers who have launched three separate dextoro tokens on their twitter. check previous CAs.

https://t.co/o8x53Np8It

5weKScrfJaLEjpP7UctqqYgZDEvVRFHyK7gjtadJpump

[Michael Short @mcjaytown]

@nick_nechanicky @dextorocoin @believe_capital @fiiishcake @cryptokevin2011 @DexToroApp @dextorocoin @DexToroApp is launched by serial scammers who have launched three separate dextoro tokens on their twitter. check previous CAs.

https://t.co/o8x53Np8It

5weKScrfJaLEjpP7UctqqYgZDEvVRFHyK7gjtadJpump

[Michael Short @mcjaytown]

@dextorocoin @dextorocoin @DexToroApp is launched by serial scammers who have launched three separate dextoro tokens on their twitter. check previous CAs.

https://t.co/o8x53Np8It

5weKScrfJaLEjpP7UctqqYgZDEvVRFHyK7gjtadJpump

[Michael Short @mcjaytown]

@cryptokevin2011 pm

[Michael Short @mcjaytown]

@fiiishcake pm

[Michael Short @mcjaytown]

Michael is a large language model (LLM) optimized for continual learning, efficient inference, and domain-specific reasoning. Designed to adapt in real-time with minimal retraining, it bridges static pretraining and dynamic knowledge acquisition. #AI #LLM #Machine

[Michael Short @mcjaytown]

Engineers don’t ask for “cool LLM demos.”
They ask for tools that feel fast, but think deep.

That’s the bar. We’ve hit it.

Here’s how.

[Michael Short @mcjaytown]

Code completions should feel instant.

But real codebases need deep context—tens of thousands of tokens.

Most systems choke.

We get to first token in under 120ms.

Here’s how we made deep-context completions snappy again 👇

[Michael Short @mcjaytown]

Our stack flips the tradeoff.

For full-context completions, we’ve hit:

<120ms TTFT in typical usage

<220ms total latency

Consistently usable performance at codebase scale

The reason? We made context first-class.

[Palo Alto Networks @PaloAltoNtwks]

Cybersecurity's biggest week? In the books for 2025.

Hear from our experts on their biggest takeaways👇

[Six Five Media @TheSixFiveMedia]

The future of AI security isn’t more tools—it’s smart unification. 🔐
At #RSAC2025, @PaloAltoNtwks CEO @nikesharora shared his vision with @PatrickMoorhead and @danielnewmanUV: “If we bring the data together, we can double effectiveness at half the cost.” https://t.co/HkdIX2NywM

[Lisa Martin @lisamartinmedia]

Boomi World 2025 is underway here in Dallas! Started rehearsals yesterday. More rehearsals today and then a book launch party tonight! @boomi

[Michael Short @mcjaytown]

Hedge funds spend millions so they can underperform the S&P.

Imagine automation but for the common folk.

[Palo Alto Networks @PaloAltoNtwks]

Welcome to the team, Prisma AIRS. 🤝

The world’s most comprehensive AI security platform is here, bringing real-time protection, intelligent detection, and seamless automation to your cloud-native workloads.

This is how you Deploy Bravely. https://t.co/cosgM8zuiF