Doug Petkanics
@petkanics
Followers
6K
Following
4K
Media
174
Statuses
4K
Doug Petkanics by day and dob.eth by night. Building web3's video and realtime AI infrastructure @Livepeer.
NYC
Joined June 2008
C’mon down to DePIN Day
Buenos Aires, today’s the day! 🇦🇷 DePINDay lands in LatAm for the very first time 🎉 We kick off at 10:00 AM — take a look at today’s agenda and see what we’ve got lined up for you ⚡️ 11:00 — 11:10 | Opening talk by @TheTomTrow { @fluence_project } 11:10 — 11:30 | @mala {
2
1
14
The week World Models went mainstream.
entropytown.com
Gaussian splats, SIMA 2, JEPA and Genie 3 — and why “world model” now means three very different things at once.
0
7
12
Livepeer is well positioned to be the leading infrastructure here, offering low latency video centric AI compute, open access, and cost-effective scalability. Read on for my vision as to why and how we need to capture this opportunity! https://t.co/epg0reNfoM
blog.livepeer.org
For the past year, the Livepeer Ecosystem has been guided by the Cascade vision: a path to transition from a pure streaming and transcoding infrastructure, to an infrastructure that could succeed at...
0
0
1
As an AI infrastructure for real-time video and generative world models, Livepeer’s open-access, low-latency network of GPUs will be the best compute solution for cutting edge AI video workflows. This category requires tremendous compute, and unlocks use cases across industries.
1
4
8
The concept of live video is expanding well beyond a single stream of video captured from a camera. Now entire worlds and scenes are generated in real-time via AI assistance. As an open video infrastructure, it's time for a vision update for what the @Livepeer Network offers...
3
10
29
Last month, something special happened in the heart of Portugal: builders, core contributors, and ecosystem visionaries came together for two action-packed days dedicated to moving the Livepeer network forward.
3
6
35
If you're new to the coming concept of World Models, give this piece by @drfeifei a read. These models will be generative, multimodal, and interactive. They'll support breakthroughs in robotics, science, and many fields of entertainment such as interactive gaming.
AI’s next frontier is Spatial Intelligence, a technology that will turn seeing into reasoning, perception into action, and imagination into creation. But what is it? Why does it matter? How do we build it? And how can we use it? Today, I want to share with you my thoughts on
0
3
10
Pack your bags ✈️ 🇦🇷 🥩🟢🎥 Breakout discussions, lightnight talks (no shilling), post-production zine, cheeky happy hour. Presented by @Livepeer, curated by @refraction_irl. Huge shout out to @kaitlynmdavies, @nick_hollins, and @misslizzyart for making this a reality 🔥
We're hosting AI x Open Media Forum in collaboration with @refraction_irl. Gathering with creatives, technologists, and researchers. You're invited to join for conversations. There will be no panels. See you in Buenos Aires.
0
4
12
More real-time AI models...Go Play!
Introducing the latest Daydream release. More open-source real-time AI models. Deeper integration with TouchDesigner (@1null1). More control over your creativity.
0
8
18
Who’s coming to Buenos Aires? We’re gathering to talk AI, video, media, and opportunities in the scene Join us for collaborative sessions, with a post-event output of a zine. Aiming for something special. Presented by @Livepeer in collab with @refraction_irl
4
6
19
Introducing @krea_ai Realtime 14B, now running inside Daydream Scope. ✨ Type a prompt, watch it instantly come alive as interactive AI video. 📼 Update prompts live. Remix forever. All open-source. ⚡ Compatible RTX 5090 and above.
5
10
54
"What does it mean to play with video generation as replayable and remixable world exploration?" https://t.co/YY9aLP2b2t
1
3
12
Big new feature release in Scope, allowing you to export, share, and remix your prompt and setting timelines to craft killer AI videos. Check it out.
A new way to control real-time video gen in Scope: - Pause at decision points, resume when ready - Track settings and prompts over time in the timeline for import/export (shareable file!) - Replay a generation and remix timeline in real-time Like your own "director's cut" for a
2
5
21
Big props to @helium for leading the charge on DePIN fee generation. It takes years to lay the infrastructure groundwork, productize, and go to market. DePIN is for the builders with endurance. (slide credit to @a16zcrypto)
0
3
19
Livepeer hits all-time-highs in Q3 🚀 Q3 2025 marked an ATH of +76% in fees and +94% in network usage, driven by real-time AI video compute processing +1M minutes per day. Decentralized infrastructure is now powering the next era of AI-driven media. Read Messari’s State of
State of @Livepeer Q3 Key Update: Livepeer's Daydream API launch propels network into real-time AI video processing, driving explosive growth in usage and revenue. QoQ Metrics 📊 • Total processed minutes 🔼 94% • AI-driven fees 🔼 131% • Total staking rewards 🔼 30% Read
7
22
66
The @Livepeer network has transformed itself into a key GPU infrastructure powering the future of AI video. Last quarter fees rose 76% to an all time high, and now 72% of the fees derive from AI video processing. Check out the full @MessariCrypto report.
State of @Livepeer Q3 Key Update: Livepeer's Daydream API launch propels network into real-time AI video processing, driving explosive growth in usage and revenue. QoQ Metrics 📊 • Total processed minutes 🔼 94% • AI-driven fees 🔼 131% • Total staking rewards 🔼 30% Read
0
4
20
Next Monday's Realtime AI Video Summit during open source AI week in SF is going to be looked back upon in a couple years as a seminal event early in the maturity of what will be a huge video world model market. The authors of StreamDiffusion, StreamV2V, and Self-Forcing amongst
3
9
25
I gave a talk at this week's @Livepeer summit highlighting the history of innovations that the project has spurred in the decentralized tech space over the last decade, all while powering millions of minutes of video processing per week for global streaming and AI applications:
3
15
43
A full day of intense hacking in Lisbon 🇵🇹 What will they BUIDL 👀
2
11
40
Excited to support StreamDiffusionV2 in the latest release of Scope today! As a reminder, Scope is tool for running and customizing real-time, interactive generative AI pipelines and models. Here is a walkthrough of it in action in Scope (see next post if you just want the
🥳We’re releasing StreamDiffusionV2 for the live-stream community—from individual creators with one GPU to enterprise platforms with many. StreamDiffusionV2 is our follow-up to StreamDiffusion: #StreamDiffusion powered real products, but temporal consistency still bugged us.
6
15
24