rightnowai.co

mx

smtp.google.com

ns

walk.ns.cloudflare.com

kianchau.ns.cloudflare.com

caa

comodoca.com

ssl.com

sectigo.com

pki.goog; cansignhttpexchanges=yes

sectigo.com

globalsign.com

pki.goog; cansignhttpexchanges=yes

letsencrypt.org

txt

v=spf1 include:_spf.google.com ~all

compai-domain-verification=org_694b183a4b2a56afeacec70c

google-site-verification=PRaSjUjpvUJefiysorK4v0sWWfTonS5FBOklSynvOAM

MS=ms30610889

dmarc

v=DMARC1; p=quarantine; pct=100; rua=mailto:e612407a9dbf4ae5b7c107439b2e6bcb@dmarc-reports.cloudflare.net,mailto:jaber@rightnowai.co; ruf=mailto:jaber@rightnowai.co; sp=quarantine

User-Agent: *
Allow: /
Disallow: /api/
Allow: /api/og
Disallow: /dashboard
Disallow: /auth
Disallow: /oauth
Content-Signal: search=yes, ai-input=yes, ai-train=yes

Sitemap: https://www.rightnowai.co/sitemap.xml

# RightNow AI

> RightNow AI is a YC-backed GPU research lab.

## Research

- [AutoMegaKernel: A Statically-Checked Agent Harness for Self-Retargeting Megakernel Synthesis](https://arxiv.org/abs/2606.09682): Statically-checked agent harness for self-retargeting megakernel synthesis. Jaber Jaber and Osama Jaber, 2026.
- [AutoKernel: Autonomous GPU Kernel Optimization via Iterative Agent-Driven Search](https://arxiv.org/abs/2603.21331): Autonomous GPU kernel optimization via iterative agent-driven search. Jaber Jaber and Osama Jaber, 2026. Code: https://github.com/RightNow-AI/autokernel
- [StreamIndex: Memory-Bounded Compressed Sparse Attention via Streaming Top-k](https://arxiv.org/abs/2605.02568): Memory-bounded compressed sparse attention via streaming top-k. Jaber Jaber and Osama Jaber, 2026. Code: https://github.com/RightNow-AI/StreamIndex
- [Ouroboros: Dynamic Weight Generation for Recursive Transformers via Input-Conditioned LoRA Modulation](https://arxiv.org/abs/2604.02051): Dynamic weight generation for recursive transformers via input-conditioned LoRA modulation. Jaber Jaber and Osama Jaber, 2026.
- [HCLSM: Hierarchical Causal Latent State Machines for Object-Centric World Modeling](https://arxiv.org/abs/2603.29090): Hierarchical causal latent state machines for object-centric world modeling. Jaber Jaber and Osama Jaber, 2026.
- [TIDE: Token-Informed Depth Execution for Per-Token Early Exit in LLM Inference](https://arxiv.org/abs/2603.21365): Token-informed depth execution for per-token early exit in LLM inference. Jaber Jaber and Osama Jaber, 2026.
- [RightNow Research](https://www.rightnowai.co/): The lab's six papers, their authors and figures, with available code links.

## Company

- [Contact](https://www.rightnowai.co/contact): Ask us anything about RightNow AI.
- [Blog](https://www.rightnowai.co/blog): Notes from RightNow AI on GPU kernels, CUDA setup and scaling laws.
- [Y Combinator profile](https://www.ycombinator.com/companies/rightnow): RightNow AI is backed by Y Combinator.

## Optional

- [RightNow Agent docs](https://www.rightnowai.co/docs): RightNow Agent, the terminal coding agent, is paused; its guides stay online as reference.
- [RunInfra](https://runinfra.ai/?utm_source=rightnowai.co&utm_medium=referral&utm_content=llms-products): RightNow AI's open models platform.

Contact: mailto:jaber@rightnowai.co
Contact: https://www.rightnowai.co/contact
Expires: 2027-09-20T00:00:00Z
Preferred-Languages: en
Canonical: https://www.rightnowai.co/.well-known/security.txt