Kevin Huang

Founding Eng @Hypercubic, prev @cognition

[Kevin Huang @SunnyKev1n]

@cl571128 Thanks Sam!

[Kevin Huang @SunnyKev1n]

Really proud of what we've built!

Hopper brings everything you need for mainframe development into one environment, with AI agents that can understand the green screen, investigate failures automatically and execute operations directly on the mainframe.

Whether you're just getting started or have been working on mainframes for years, Hopper's customizable layouts let you work the way you want.

Excited for you to try it!

Download and try Hopper for free: https://t.co/A8jXHC9i3j

#COBOL #mainframe #agents

[Aayush Naik @aayush_naik]

Read the announcement here: https://t.co/yN2W2dP5l4

Download Hopper today: https://t.co/07nfgIXNCq

[Kevin Huang @SunnyKev1n]

@aayush_naik Really proud of what we’ve built! Lots more to come!

[Kevin Huang @SunnyKev1n]

@aayush_naik It looks awesome!

[Kevin Huang @SunnyKev1n]

We’re growing the team and currently hiring Founding Engineers and Software Engineer Interns.

Apply here: https://t.co/OErrTu45xd

[Kevin Huang @SunnyKev1n]

Proud to share that Hypercubic has raised a $5.3M seed round, led by @civ_co!

We’re building an AI-native platform that helps enterprises understand, modernize, and verify critical mainframe systems.

As a founding engineer, I get to build on my experience with agentic products while exploring what they can make possible for mainframe modernization. That means taking new ideas from zero to one and shape a new generation of tools for understanding, transforming, and verifying these critical systems.

Working at Hypercubic alongside Sai Gurrapu, Aayush Naik and the rest of the team has been some of the most technically challenging and rewarding work of my career. With this round, we’re just getting started.

We’re looking for engineers who think from first principles, are comfortable with mathematical reasoning, and want to solve hard problems with real-world impact. If that's you, come join us. See open roles in comments.

[Sai @saigurrapu18]

Today, we’re announcing our $5.3M seed round, led by CIV.

There is a rare window right now to redefine how some of the world’s most legacy software gets rebuilt across banks, airlines, governments, and large enterprises.

For decades, mainframe modernization has been a massive human undertaking: multi-year programs, tens of millions of dollars, and large teams reconstructing, rebuilding, and verifying systems that nobody fully understands anymore.

We started Hypercubic with a simple thesis: mainframe modernization should become a repeatable, software-driven process rather than a bespoke human undertaking.

Over the past year, we’ve made several breakthroughs in an attempt to build the most capable mainframe modernization platform one that can understand, rebuild, and verify critical systems at high velocity.

We’re already seeing with customers how dramatically parts of the modernization lifecycle can be compressed. The industry is not used to the speed we now know is possible.

Today, this work remains fundamentally a human labor problem. We think it can become a computational problem instead.

@aayush_naik and I are building a small, high-caliber team to go after this. If this problem excites you, please reach out, we’re hiring exceptional people across the team!

[Kevin Huang @SunnyKev1n]

HyperLoop is our verification-driven system for mainframe modernization.

Mainframes still run some of the most critical software in the world. Modernizing them is necessary, but it cannot be a leap of faith.

HyperLoop combines AI-powered development with layered verification, so enterprises can move critical systems forward with confidence.

#mainframe #COBOL #modernization

[Aayush Naik @aayush_naik]

Today we’re announcing HyperLoop, a verification-driven system for mainframe modernization.

HyperLoop marks the culmination of everything we’ve been working on since starting Hypercubic.

[Kevin Huang @SunnyKev1n]

@aayush_naik AI can make modernization faster but verification is what gives enterprises the confidence to move critical systems forward. Excited to keep pushing toward a future where even the most critical legacy systems can evolve safely!

[Kevin Huang @SunnyKev1n]

@aayush_naik Can’t wait to see where this goes

[Kevin Huang @SunnyKev1n]

There’s 200x more COBOL transactions than Google searches everyday.

The tooling around them has barely moved either. Most mainframe developers still work the way their predecessors did in the 1970s with green-screen terminals, ISPF panels, function keys, dataset names.

Hopper was built for the engineers running the systems too critical to ever fail. A real TN3270 terminal with an agent that helps draft JCL, navigate ISPF, walk JES spool, debug return codes and keeps the engineer in the loop.

You drive. The agent does the work.

Download Hopper today: https://t.co/A8jXHC9i3j

#COBOL #mainframe #TN3270 #IBM

[Sai @saigurrapu18]

Today we’re launching Hopper — the first agentic development environment for the mainframe.

Modern AI agents can inspect code, edit, test, and debug but they’ve never been able to touch the systems that actually run the world: core banking, payments, insurance, airlines, government.

Until now.

Hopper combines a real TN3270 terminal, native mainframe panels, and a powerful AI agent that can navigate ISPF, inspect datasets, write JCL, submit jobs, parse spool output, query VSAM, talk to CICS, and explain failures.

Once AI agents can operate inside the mainframe environment, maintenance and modernization changes.

At Hypercubic, we are building AI-native infrastructure for the full lifecycle of legacy modernization.

Hopper is our first step toward bringing agentic development to the most critical computing environments in the world.

[Kevin Huang @SunnyKev1n]

My favorite moment from building this was watching the agent go through a JCL workflow batch and navigating the TN3270 terminal on its own to investigate an issue and asked me if it should patch and resubmit, all in a few minutes. A mainframe developer using Hopper told us the same task would've taken him hours.

That's when it switched from feeling like a prototype to feeling like a real product. Excited to keep building Hopper with @saigurrapu18 @aayush_naik 🚀

[Devin Developers @devindevelopers]

GPT-5.1 and GPT-5.1-Codex are now live in Windsurf

All paid users get free access for the next 7 days. GPT-5.1 is also set as the default model for new users.

[Alvaro Cintas @dr_cintas]

Cursor and Windsurf both launched new models.

So I decided to put them to the test in speed, up to date info, and app creation.

Here are the full breakdown & results:

(Video is not sped up 🤯)

[Kevin Huang @SunnyKev1n]

Ship faster than ever before with the new SWE-1.5 model!

[Cognition @cognition]

Today we’re releasing SWE-1.5, our fast agent model.

It achieves near-SOTA coding performance while setting a new standard for speed. Now available in @windsurf.

[Devin Developers @devindevelopers]

You can now try a new stealth model in Windsurf.

Falcon Alpha is a powerful agentic model designed for speed.

We're excited to hear what you think ⚡️

[Kevin Huang @SunnyKev1n]

What a ride so far. Super pumped for what’s ahead!

[Cognition @cognition]

We’ve raised over $400M at a $10.2B post-money valuation to advance the frontier of AI coding agents.

The round was led by Founders Fund with other existing investors including Lux, 8VC, Neo, Elad Gil, Definition Capital, and Swish VC all doubling down. We’re also joined by new investors including Bain Capital Ventures and D1 Capital.

Two of our early investors, Christian Lawless of Conversion Capital and Emily Cohen of Neo, have even joined our team full-time.

[Kevin Huang @SunnyKev1n]

@cognition Let's go! Super excited for the future

[Kevin Huang @SunnyKev1n]

@BlogTriggers @itsandrewgao @windsurf You can toggle this in the Windsurf settings

[Kevin Huang @SunnyKev1n]

@FernandesS63056 @itsandrewgao @windsurf You can update the rules file with this template from https://t.co/zqygCuP1wK under "New Coder" which makes the model act as a tutor to help you learn.

[Kevin Huang @SunnyKev1n]

@FNCINDRA @itsandrewgao @windsurf You can already do this by highlighting the lines of code and pressing CMD + L (or CTRL + L on Windows). It will @ mention the lines of code highlighted directly in the Cascade prompt.