The New Primitives: Building AI Native Software — Kwindla Kramer, Daily

AI Engineer · 21 min · 247 sentences · from YouTube's caption track

Each timecode opens YouTube at the start of that sentence. Line anchors (#s42) are the cue ids in the WebVTT, and every line carries its start and end seconds. All transcripts has every talk, and the whole corpus as one file.

00:01[music]
00:12Good morning.
00:14I know a lot of you in this room.
00:15It's great to see you.
00:17Welcome to the voice track at AI Engineer World's Fair.
00:20For those of you who don't know me, my name is Quinn La Holman Cramer.
00:24I work at a company called Daily.
00:26We make developer infrastructure for real-time audio, video, and AI.
00:31And we're the team behind Pipe Cat, which is the most widely used framework for building voice agents today.
00:37Pipe Cat is open source and vendor neutral.
00:39It's used by companies like AWS and Nvidia and Anthropic and thousands of startups and scale-ups and enterprises.
00:48And today I'm going to talk about what kind of agents we're building today, including voice agents, but not just voice agents, and what I'm interested in building next.
00:58And I'm going to try to put all this in the context of the roughly 80-year history of digital computing so far.
01:06So, we've got a lot to cover.
01:07We're going to go fast.
01:09But we're going to start in 1945 with an essay called As We May Think, written by an engineer, an academic, a civil servant named Vannevar Bush.
01:19Bush deeply understood technologies ranging from analog computers to photography to radio to radar.
01:26As We May Think is a extraordinary piece of writing.
01:31The essay predicts the development of, among other things, document display on a screen and document scanning and OCR and speech-to-text and text-to-speech and programming languages and hypertext and search engines and data networks.
01:45Something like the GoPro camera, something weirdly like the Amazon Kindle store, and voice interfaces and brain computer interfaces.
01:53And I've been thinking a lot about As We May Think lately because Bush wrote this essay right at the very beginning of the computing age.
02:03And I think it feels to most of us like we're working right at the beginning of a new age, the intelligence age.
02:09So, what will we build?
02:13Well, at the moment we're building agents and we're having a lot of fun doing it.
02:19And a lot of the AI engineering work we're all talking about this week is focused on building a full coherent software stack for AI agents.
02:28Here is Satya Nadella talking a couple weeks ago on a crossover episode of the No Priors and Latent Space Pod about the challenges of building agents in 2026.
02:38That's sort of
02:39That's right.
02:39So, so in some sense you kind of want to harness to define the models, the the data, uh and the tools.
02:49And so that you have a loop across those three.
02:51And so what we are trying to first of all make sure is each of our products that we build, right?
02:56Whether it's GitHub Copilot or the security copilot the stuff we showed with M dash or even the discovery for science, it doesn't matter.
03:04All of them are multimodal harnesses um with tools access so that you can do this progressive uh disclosure of tools even so that they're token efficient.
03:14Uh and then you're feeding it with very rich context.
03:20So, if you were here last year at AI Engineer World's Fair, you could draw a through line from the things we were talking about last year to loops and tool calls and context engineering and the stuff we're focused on this year
03:31to some emerging ideas.
03:33Uh you can hear that in Nadella's clip just there.
03:37I think of this is kind of agents plus plus, like multi-model harnesses and software copilot embedded in every single piece of software, organization level harnesses.
03:47So, how do we go from agents to agents plus plus to the next thing beyond agents.
03:54Well, the last time we had this kind of massive change in how we write software and what we write software for and to do was the early days of the World Wide Web.
04:03And I was around for the early days of the World Wide Web.
04:06I was a baby programmer in 1995 and the thing we talked about all the time in 1995, the way we talk about agents today, is web pages.
04:15I spent a lot of time writing HTML by hand and building web server software in C and indexing and search software in C and authoring tooling and management infrastructure for web pages in Pearl.
04:27I was as excited about HTML in 1995 as I am about agents today.
04:34And the web page is still with us and it's still important and useful.
04:37But today we talk a lot more about web applications and native mobile applications than we talk about web pages.
04:44So, just like we went from web pages to full-blown web and native mobile, clearly we're going to chart a path to a new fully AI native software that comes after agents and agents plus plus.
04:57So, let's keep going back in time to in order to think about this future.
05:02Here's a timeline Vannevar Bush lays out and as we may think, he talks about the abacus, which was both an immensely useful device for doing practical everyday mathematical calculations
05:14and also an incredibly important theoretical tool that led to ideas like numeric place value and the concept of zero.
05:22And Bush talks about the massive jump from the abacus to the state-of-the-art electromechanical keyboard calculating machines that he had in 1945.
05:33And then he posits that we're about or he is about to witness and help create an equally large leap to what he calls the arithmetical machine.
05:48And then he goes a step even further than that and he invents or designs in that essay a device he calls the memex.
05:56And we have a little bit of an advantage over over Bush in 1945.
05:59We've seen 80 years of computing play out.
06:02So, we can modify his timeline a little bit.
06:04We can go from the abacus to the stored program computer to 40 years later the personal computer and 40 years after that this AI agents era that we're all collectively
06:15helping to invent and create and bring into being.
06:18So, the question for me is what did we build to go from those very first digital computers in the 1940s to the personal computer in the 1980s?
06:28Well, in the 1950s the big job was to figure out more effective ways of transmitting human intent to these new computing machines.
06:36We built the first programming languages.
06:37We wrote the first compilers.
06:40And the the theoretical underpinnings here were figuring out how to combine the elegance of mathematical formalisms with something a little bit more like natural language.
06:51And then building on that in the 1960s the challenge was to make these machines interactive.
06:56Make these machines capable of a two-way dialogue with humans.
07:02The '60s also saw the birth of graphical programming with systems like Ivan Sutherland's Sketchpad.
07:09And the '60s were an amazing era for science fiction.
07:12Even though almost nobody had access to a computer the computer became a big part of the popular imagination.
07:19The idea of a computer really resonated with people and ideas matter.
07:24For example, here is the idea of the computer in Star Trek.
07:29Put her on record.
07:31Recording.
07:34Come.
07:36Captain's log supplemental.
07:39Engineering officer Scott informs warp engines Can be made operational and re-energized.
07:45Computed and recorded, dear.
07:49Computer, you will not address me in that manner.
07:51Computer.
07:53Computed, dear.
08:00I love the background sound of punch cards going through a punch card reader.
08:04So, like you know the computer is working even though it's talking to you about what it's actually computing.
08:11There were of course a bunch of other talking computers in in science fiction of the '60s and the next year after this, the Kubrick movie that was a interpretation of Arthur C. Clarke's 2001: A Space Odyssey had the HAL 9000 computer.
08:22This is a much, much more dystopian view of a talking computer than the Star Trek computers.
08:29And by the 1970s, computers had become powerful enough that the next big job was designing abstractions that could scale to much larger amounts of data and much more powerful
08:38computing substrates.
08:40We got relational databases, which introduced new theoretical underpinnings for data manipulation.
08:46And we got declarative languages, which leveraged those new theoretical insights.
08:50And programming languages in general continued to evolve in what to me at least are really amazing ways.
08:55We got Smalltalk and object-oriented programming in the '70s.
08:59And all of this set the stage for the personal computer in the 1980s.
09:03The Macintosh shipped in 1984.
09:05Windows 1.0 shipped in 1985.
09:08And Microsoft's mission statement was a computer on every desk and in every home.
09:12And incredibly, Microsoft delivered on that mission statement.
09:17And we got a computer on every desk and in every home because these new personal computers delivered real, amazing, tangible benefits.
09:24Take VisiCalc, for example, which was the first spreadsheet program.
09:29A truly new abstraction for doing computation, numerical computing, two-dimensional, interactive, so durable and so useful that probably most of us in this room use a direct descendant of VisiCalc
09:42regularly, Google Google Sheets or Microsoft Excel or whatever.
09:49Or put another way, this was a spreadsheet in 1957.
09:54And this was a spreadsheet in 1985.
09:57And I think a lot about VisiCalc these days too because I think VisiCalc is an example of how transformative new technologies can be in the way of delivering a capability
10:06that used to require a lot of specialized people and specialized knowledge and making it generally accessible.
10:11And I think VisiCalc is a potentially a counter-argument to the argument or the fear or the concern that AI is going to lead to mass unemployment.
10:20Because VisiCalc didn't put accountants out of business.
10:22Instead, it made much much much much more accounting-like work possible.
10:27And it made new categories of work possible that we couldn't even really conceive of when a spreadsheet or doing a screen's worth of calculations as we think about it today
10:38took a roomful of people.
10:41So if we were here in the Moscone Center in 1985 and these two interfaces, the Macintosh System 2 and Windows 1.0 were state of the art, what would we have said the world would look like in 10 or 20
10:54or 30 or 40 years?
10:57Well, we actually have a really great example of a prediction from that time.
11:01Like as we may think another famous document in the history of human computer interaction, concept video from Apple made in 1987 called Knowledge Navigator.
11:10This is very much worth tracking down online and watching all of if you haven't seen it.
11:14I'm just going to play about 20 seconds from the middle.
11:22You have three messages.
11:24Your graduate research team in Guatemala, just checking in.
11:28Robert Jordan, a second semester junior, requesting a second extension on his term paper and your mother reminding you about your father's
11:36surprise birthday party next Sunday.
11:40So, the video shows a foldable tablet, a touchscreen interface, [clears throat] a conversational voice assistant with a really strong personality, access to both global and personal information, real-time video generation, real-time computer vision, seamless video call integration,
11:55delegation of complex tasks for autonomous execution, and what we might call today continual learning.
12:02And it's really, really clearly influenced by AS we may think, but it's also quite different.
12:06It really is updated for 40 years of progress, and it really does sort of presage this AI agent era we're in now in a way that Vannevar Bush's Memex didn't and maybe couldn't.
12:17The Knowledge Navigator video divides our timeline, I think, quite neatly in half, and hold that thought cuz we're going to come back to it.
12:25The 1990s were about the network, first local area networks and dial-up, and then the internet and the web, and with the benefit of hindsight, I now think that the single
12:37most important thing about the web was that it was multimodal from the very beginning.
12:41More even than the GUIs of the 1980s, the web anticipated that text and audio and video and data were not different things to be used in different programs, they belonged
12:52together.
12:53And in a real sense, the web was an attempt, and a conscious attempt on the part of a lot of people building the web to make that Knowledge Navigator video real.
13:03Then in the first decade of the new millennium, the big job was to make all of this computing stuff mobile and continually connected, to put this new multimodal networked computer
13:14in your pocket, literally, to give a supercomputer to everybody in the world that they could carry around in their hand.
13:21And as with the 1960s, there was an efflorescence of like futurism on screen in the first few years of the new millennium.
13:29And I think it was because computers you could carry around with you and cameras everywhere and a kind of Moore's law for pixels making screens super cheap really gave us a chance
13:41to think through what we thought the future would look like in a new way.
13:45A lot of stuff we could almost but not quite build was cohering in the minds of people working on these machines.
13:52And the best and most famous Hollywood computers from that era were created by John Underkoffler [clears throat] for the films Minority Report and Iron Man.
14:00Here's Minority Report from 2002.
14:05It's no longer there.
14:07Time frame?
14:0813 minutes.
14:09Hey Chief, investigator from the Feds here.
14:12Yeah, I don't need some twink from the Fed poking around right now.
14:14John, I wrote it down on your calendar.
14:15I left you a message at your house.
14:16Check in with the Favors ahead of Ford and see if the neighbors knew where they went.
14:18Check all relations.
14:19Check the neighbors and relations.
14:20But John
14:21It's actually
14:21Just get him some coffee.
14:22Tell him some stories how I save
14:23[music]
14:23your ass every day and you can't do without me.
14:25I got coffee.
14:26Thank you.
14:27Danny Witwer.
14:28Twink from the Fed.
14:29Oops.
14:30Gone.
14:32So the gestural interface in Minority Report was implemented on screen as special effects, but it was actually based on John's PhD work at the MIT Media Lab.
14:39In a real sense, this was real technology.
14:42John had brought the UI out of the small screen and into the world with us in a bunch of really interesting and lovely ways.
14:48John also consulted on Iron Man, which is a very different view of the future than you Minority Report, which was Spielberg working in like the American Kubrick dystopian tradition.
14:56Iron Man is really squarely in that Star Trek goofy futurist tradition.
15:00But I think you could see the common elements in the UI depicted on screen.
15:05It's still from the same era.
15:12Wake up, Daddy Sean.
15:13Welcome home, sir.
15:14[music]
15:15Congratulations on the opening ceremonies.
15:17They were such a success.
15:18As was your summit hearing.
15:20[music]
15:20And may I say how refreshing it is to finally you in a video with your clothing on, sir.
15:27You!
15:29I swear to god, I'll dismantle you or short your motherboard.
15:31I'll turn you into a wine rack.
15:37I co-founded a startup with John in 2006 to make the Minority Report interface into a commercial product.
15:43This is our demo reel from 2012, 6 years into that work.
15:49[music]
15:53[music]
17:10This long project to build the multi-modal, multi-device, multi-screen, multi-player, ubiquitously connected computer is still what I'm working on 15 years later.
17:20In 2010s, we built out the cloud, which laid the groundwork for the infrastructure and data centers and data capacity we would need to scale up AI training and inference,
17:30which brings us to now.
17:33We're building agents.
17:34And we're starting to think about agents plus plus.
17:38But I think we can also start to think about the next thing, the AI native software that is to agents what today's internet is to the web pages of 1995.
17:47And one way to think about the story is this.
17:49We went from the calculator to the computer to the personal computer to the global cloud computer.
17:54And now we actually have the ability to build the Memex and Jarvis from Iron Man and knowledge navigator from that 1987 video for real, completely working.
18:04And by building those things, we'll figure out what we want to build next.
18:09A couple of weeks ago the team at Tavis released a reimagined knowledge navigator video this time entirely built on real and available technology.
18:17I'll just play another 20 seconds of this, but like the original knowledge navigator video, this is worth tracking down and watching in full.
18:31Good evening, Hassaan.
18:32I adore that houndstooth jacket you're wearing today.
18:35Anything I can help with or would you like to review tomorrow's schedule?
18:38Thanks so much, Tom.
18:40Yeah, let's review tomorrow's schedule and see how busy it is.
18:44Opening your calendar now.
18:45Here is the quick version since it is late.
18:48Tomorrow morning is slammed.
18:50Investor meeting at 9:00, then back-to-back one-on-ones and internal meetings until 1:00.
18:55So, the full 4-minute video is one take, completely real.
18:59And when you watch it, it really does feel both like the knowledge navigator video from 1987, familiar but built on real technology, and like something brand new.
19:09And I'll just close with a massively multiplayer game project I've been working on with some friends as a canvas to really think about what AI native software can be.
19:20This game is built from the ground up with LLMs as the core of every interaction.
19:25At every moment in the game, there are hundreds of inference calls happening and we couldn't have built anything like this even a year ago.
19:32Oh, sorry.
19:36Welcome to Gradient Bang, a multiplayer game
19:39[music]
19:39that showcases real-time agent orchestration.
19:42Gradient Bang demonstrates several patterns for AI sub agents such as asynchronous non-blocking context compression.
19:50Okay, make a note for later.
19:51We are going to eliminate Heliotrope from existence.
19:54Noted.
19:55Long-running sub agents that share context.
19:57[music]
19:58Eagle is on five trade loops.
20:00Hawk and Raptor are on five exploration loops
20:03[music]
20:04each.
20:04Your fleet is busy.
20:06Progressive skills loading.
20:08How much does your average ship cost?
20:10Ships range quite
20:11[music]
20:11a bit, Captain.
20:12Dynamic user interface generation.
20:14Show my task history.
20:16Certainly.
20:17Uh hide the map.
20:19Okay.
20:20And conversational
20:20[music]
20:21voice.
20:21No, I don't want to exchange it.
20:23I just want to sell it for cold, hard cash, please.
20:26I'm afraid
20:26[music]
20:27the galaxy doesn't allow you to be shipless and hitchhike.
20:31So, I went long after wrap-up, but I will say that the first version of this new draft talk was an hour.
20:37So, I have a lot more things I'm super excited to talk about with all of you.
20:41So, if you are interested in this stuff, come find me.
20:44We have a booth on the show floor.
20:45I'm online everywhere and I'm excited to build agents cuz agents are awesome, but also to build the next next thing, too.
20:53Thank you.
20:54[applause]
21:12[music]
21:13Mhm.