Subham Kundu

Get the shit done. Founder, Backed by @YCombinator

[Subham Kundu @subhamCenrax]

@Govikavaturi @AkonLabs @ycombinator Thanks

[Subham Kundu @subhamCenrax]

Today we're launching @akonlabs backed by @ycombinator

We build GitNexus, an open-source knowledge graph for your repository. Coding agents query it to get exactly the context a task needs symbols, call paths, dependencies instead of grepping around and stuffing half your codebase into a context window.

Context is the bottleneck for coding agents. Bigger windows haven't fixed it. Buyers of agents are paying more per token to get worse code, and the agents themselves have no map of the repo they're editing.

We're fixing this. On DeepSWE, the same agent with GitNexus vs. without:

1. 37% → 68.4% of tasks passed
2. Half the cost per fix
3. On the hardest tasks, the gap more than triples

One stack: knowledge graph, coding agent, automated PR review. Managed, self-hosted, or air-gapped.

45k GitHub stars. 65k weekly npm downloads. 10+ enterprise deployments.

Sign up at: https://t.co/8SLwkzhe7P

[Subham Kundu @subhamCenrax]

@aatharva16 @AkonLabs @ycombinator Yes we have a 14 day free trial

[Subham Kundu @subhamCenrax]

You can use Gitnexus directly now on https://t.co/0piD7GYCeQ

@claudeai @AnthropicAI @ycombinator

[Subham Kundu @subhamCenrax]

@rom1trs Os

[Render @render]

Coding agents can now go beyond writing code on Render.

🔐 Connect via MCP OAuth
🚀 Trigger deploys & change service plans
🧠 Tap into 20+ Render Skills
⌨️ Manage services, Postgres & Key Value with the CLI

Build, deploy, and operate - read the announcement → https://t.co/UQomO7jROA

[Render @render]

GitNexus, the open-source code intelligence platform by Akon Labs, now deploys to Render in one click.

Get a live instance in minutes with no config or infra setup → https://t.co/L9X8X7CAYP

[Subham Kundu @subhamCenrax]

@noahelhadedy I built the open source version of it that runs both on mac and windows https://t.co/YpfbBLFpJ0

[Subham Kundu @subhamCenrax]

@taoleeh There is a open source version of it: https://t.co/YpfbBLFpJ0

[Subham Kundu @subhamCenrax]

@Spywrites I built the open source version of it

https://t.co/YpfbBLFpJ0

[Subham Kundu @subhamCenrax]

@SpotifyEng @ClaudeDevs @GeminiApp @OpenAI @Spotify If anyone wants to try out the open source version that runs both on mac and windows do have a look: https://t.co/YpfbBLFpJ0

[Subham Kundu @subhamCenrax]

@avibose22 @GoogleDeepMind Congrats @avibose22

[Subham Kundu @subhamCenrax]

@AlfieJCarter Agents

[Subham Kundu @subhamCenrax]

@alokbishoyi97 Add me

[Subham Kundu @subhamCenrax]

RLMs and loop engineering are the same bet in different clothes.

RLMs: don't cram the context into one call, recurse over it.

Loop eng: don't cram the task into one prompt, loop around it.

The unit of intelligence isn't the prompt anymore. It's the loop.

[Subham Kundu @subhamCenrax]

Preparing for interview got more easier. Basics, Hands on exercise and mock interviews all in one place.

Try out: https://t.co/nV1aaAxpfE

[Sam Altman @sama]

I am happy everyone is switching to Codex, but Tibo if you start rate limiting me or making me use worse models...

[Subham Kundu @subhamCenrax]

@Yuchenj_UW Thanks @Yuchenj_UW

[Yuchen Jin @Yuchenj_UW]

OpenAI just dropped a training challenge:

Train a <16MB language model in 10 minutes on 8×H100s and minimize held-out loss on a fixed FineWeb dataset.

Basically NanoGPT Speedrun. They’re sponsoring $1M in compute.

I can summon my autoresearch army to win it… if I have time.