Dillon Pulliam

Researcher & Engineer

[Dillon Pulliam @dillonpulliam30]

This week I joined @ToyotaConnected! I'm thankful for this incredible opportunity and excited to be a part of the team!

[Dillon Pulliam @dillonpulliam30]

Thanks to everyone who supported the Parawise journey, gave encouragement, and checked in on progress, it meant a ton.

With this chapter wrapping up I'm excited for whatever comes next. If you're working on something innovative and exciting I'd love to learn more!

[Dillon Pulliam @dillonpulliam30]

On top of the business logistics and technology, I learned that starting a start-up isn't easy, attempting to bootstrap to profitability isn't easy, and trying to do it with a team of 2 where one is full-time and the other is part-time isn't easy.

(continued in next post)

[Dillon Pulliam @dillonpulliam30]

My brother (Co-Founder) and I recently decided to shut down Parawise which we created about a year ago.

Although not the outcome I imagined when beginning this adventure, it's been a year full of building and constant learning.

(continued in next post)

[Dillon Pulliam @dillonpulliam30]

My brother @zapulam and I recently founded Parawise. If you’d like to learn more about what we are building you can follow @ParawiseCO and check out our website: https://t.co/b6zYNf3YJ5 . Building!

[Greg Brockman @gdb]

When doing deep work, context switching is the enemy.

[Dillon Pulliam @dillonpulliam30]

To see the Museum's collection to date: https://t.co/uWOVAu0DJh

[Dillon Pulliam @dillonpulliam30]

This fall my wife, Maija, is partnering with @visual_museum to photograph, document, and gather evidence of women in the visual history of the Church, specifically in Spain! If you'd like to learn more, or see how you can support the work she is doing: https://t.co/v9EA0U313E

[AssemblyAI @AssemblyAI]

Applying LLMs to audio is hard.

Introducing LeMUR, short for Leveraging Large Language Models to Understand Recognized Speech.

Our new framework to apply LLMs to audio transcriptions and spoken conversations.

[Dillon Pulliam @dillonpulliam30]

Super excited to share AssemblyAI’s new Conformer-1 – a robust state-of-the-art speech recognition model!
https://t.co/mnyDGoJnpr
#AssemblyAI

[AssemblyAI @AssemblyAI]

Introducing Conformer-1: our latest state-of-the-art speech recognition model.

Built on top of the Conformer architecture and trained on 650K hours of audio data, it achieves near-human-level performance, making up to 43% fewer errors on noisy data than other ASR models.

1/9

[AssemblyAI @AssemblyAI]

We successfully processed one billion audio files!

That's 1,000,000,000 files and over 1T tokens transcribed!🤯

[AssemblyAI @AssemblyAI]

Our Founder & CEO @YouveGotFox was a guest on The Machine Learning Podcast!🎧

If you're an ML developer, PM, or founder, tune in to learn about our mission and the unique challenges of building a business with Machine Learning at the core.

[Stack Overflow @StackOverflow]

On our latest episode, we sat down with @YouveGotFox, CEO & co-founder of @AssemblyAI, to talk about what sparked his interest in AI and machine learning and how their speech-to-text solutions help major companies like Spotify and the BBC.

https://t.co/sbNkaIYhra

[AssemblyAI @AssemblyAI]

We are excited to share Light-weight probing of unsupervised representations for Reinforcement Learning.

It’s a probing protocol highly correlated with RL performance, with lower variance and up to 600x lower computational cost compared to standard RL training/evaluation.

🧵

[Dillon Pulliam @dillonpulliam30]

In our weekly Deep Learning Paper Recap we take a look at two interesting papers. To learn more about what's exciting in these papers, their key findings, and our take-aways, please feel free to check out the Paper Recap!

https://t.co/2XdHXZDMCg

[AssemblyAI @AssemblyAI]

This might be one of our most comprehensive Open Source Deep Learning projects and tutorials yet!

In this Guide @r_o_connor teaches you how to implement your own Text-To-Image model, and it includes all code on GitHub and a PyPi package!

https://t.co/1jmh5QXqXF

[Accel @Accel]

.@AssemblyAI is making complex Deep Learning technology accessible to all developers. Since we first partnered in February, their community has grown tremendously. Hundreds of product teams now use their APIs to analyze speech, transcribe audio, & more. 👏https://t.co/zPr1ImAxcs

[Dillon Pulliam @dillonpulliam30]

Exciting news for the team! 🚀🚀🚀

[AssemblyAI @AssemblyAI]

We have BIG NEWS🥳

Four months ago, we announced our $28M Series A.

Today, we're incredibly excited to share that we’ve raised another $30M in our Series B led by @insightpartners!

https://t.co/Wm3iFdu1sq

[KB @kbizzle_kb]

@AssemblyAI is #hiring a #remote Staff #Python Engineer to join our growing team! You'll help us build highly scalable, developer-centric APIs powered by state-of-the-art #ML models in #speechrecognition & #NLP! You can apply directly at https://t.co/pCpbjLRxHz!

#remotejobs