speko ai

Building Speko (YC S26). OpenRouter for voice AI.

[speko ai @spekoai_]

@aaronjmars @LiquidPadBot @_proxystudio @aaronjmars can we have a quick chat?

[speko ai @spekoai_]

@MusliSusi we didn't launch any coin but i will try to launch my own coin

[speko ai @spekoai_]

I have a question for anyone building with voice AI: https://t.co/gxSk5DyNI5
What's the most painful part of your stack right now?

- STT accuracy for non-English?
- Latency spikes in production?
- Testing voice quality across updates?
- Managing multiple provider configs?

Genuinely want to know. Building VoiceForge to solve
the biggest pain points. DMs open.

[speko ai @spekoai_]

Major benchmarks rank gemini 3.1 flash tts #2 in the world.

we asked it to read a 1507-word script for ten minutes.

353 windows of severe-tier failure.

cartesia and elevenlabs combined: 0.

a 30-second elo score is not a production benchmark.

More here: https://t.co/uuK4RRZHbW

[speko ai @spekoai_]

Got into YC interview recently and built a simulator to practice with @garrytan

Link: https://t.co/eXZD5NAonD

[mleejr @MLeeJr]

#1 👀

[speko ai @spekoai_]

The AI voice space is filled with opportunities. Future

[speko ai @spekoai_]

Turned this into a Claude Code skill so anyone can do it. Six commands: init scans your project, compile builds the wiki, query answers questions against it, lint heals inconsistencies. The LLM owns the wiki, you steer with questions.

Open sourced: https://t.co/2RR2hwXDPU

[Andrej Karpathy @karpathy]

Someone recently suggested to me that the reason OpenClaw moment was so big is because it's the first time a large group of non-technical people (who otherwise only knew AI as synonymous with ChatGPT as a website) experienced the latest agentic models.

[Andrej Karpathy @karpathy]

Judging by my tl there is a growing gap in understanding of AI capability.

The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code.

But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along.

So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions.

TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.

[speko ai @spekoai_]

@karpathy @karpathy 's article, asked CC to rebuild my project DB, looks great so far

[speko ai @spekoai_]

@karpathy @karpathy 's article, asked CC to rebuild my project DB, looks great so far

[speko ai @spekoai_]

my most common message these days to Claude Code

[speko ai @spekoai_]

Does this look proper for a software factory?