Built an Investor directory called Launchnome
- 30k+ Investors listed (more coming soon)
- 17k+ VCs
- 11k+ Angels
- 600+ Family Offices
- 250+ Grants available on https://t.co/zTbDkJIIhG
More features on the way
There is so much to share about the Momentum but our sole focus right now is Sprint S1 & to help all the founders & startups out there to build what's next.
We ourselves are building products & we know how hard it's to take the first step. Sprint helps you exactly with the first step & foundation. Our target is to pack solid 6 months in 30 days and to help you achieve your first dollar.
We will invite speakers & mentors to help you with anything, we will push you relentlessly, we will ask you questions & tell you answers, we will connect you with industry leaders and so much more. Sprint packs unmatched benefits & everyone should leverage it to achieve their dreams.
Apply - https://t.co/kxj7PVd8a6
Exactly a month ago, we announced about GIKSN Residency & today the forms are live for both GIKSN Momentum & Sprint S1. Upto 3 startups will be directly selected for Momentum through Sprint S1.
GIKSN Momentum
- Month long offline residency
- December 2026 tentatively
- India
- 15 startups
- Focused on taking these 15 startups to the next level
- Apply - https://t.co/uYW1PBxSE1
GIKSN Sprint S1
- Monthly online cohorts
- Starting on the 1st of each month
- Fully Online
- Upto 200 startups in S1
- Focused on early stage & foundational growth
- Upto 3 startups will get direct ticket to GIKSN Momentum.
- Apply - https://t.co/uYW1PBxSE1
Format & Experience coming soon.
If argentina wins today, we're going to host a residency in india powered by @giksnresearch
- developers
- founders
- researchers
- marketers
- 30 days, 15 teams, 1 villa, unlimited credits, & more
who's in?
Day 0 of building in public.
goal: grow giksn research to $10k mrr from scratch.
i'm building this in the open, focused on solving one of the biggest bottlenecks in ai today: memory and language model infrastructure.
today's ai agents are great at demos, but they still struggle with long-term memory, context, and maintaining understanding across complex workflows.
alongside the research, i'll also be building micro saas products that solve real user problems using the infrastructure we're developing.
the goal is simple: turn research into products people actually use.
what's the biggest memory or context issue you've run into with ai tools?
Rinne v0.1.7 is now live
With v0.1.7, Rinne now supports two very crucial features which are MCP & Skill integrations.
Users can now connect any MCP or Skill seamlessly.
For more details - https://t.co/d7nu7tRUPk
Rinne v0.1.7 is now live
With v0.1.7, Rinne now supports two very crucial features which are MCP & Skill integrations.
Users can now connect any MCP or Skill seamlessly.
For more details - https://t.co/d7nu7tRUPk
Want to contribute open-source, but confused where to start?
We have opened a few issues in Rinne's repository https://t.co/eYfQNyChgm with most of them being good first issue.
We are creating more issues as we go towards production-ready
Rinne has been downloaded 97 times from the day we have made it open-source. Not much but enough to keep us going & building the best multi-LLM harness.
MCP & Skills connectors are on their way and many more features are in the pipeline.
One major obstacle between now & the best harness is Memory which we are already working on!
Rinne started as a testing ground for the product we are working on (still building).
We started using it not to just test it but for building it also, & right now Rinne comes with a lot of issues & gaps. We are planning to close these issues & gaps as soon as possible.
v0.1.6 is already live and provide a very clean user experience. v0.1.7 will come within the next 48 hours which will contain everything v0.1.6 is offering alongside the new features which are
MCP integrations
Skills integrations
v0.1.8 will contain
New Conductor
Enhanced Loops
We anyway need a brutal testing ground so we decided to continue improving Rinne!
cc @thedeepanshuweb
The best AI model is a trap. The teams pulling ahead orchestrates many.
We though why not build something which can use all your available harnesses & API, optimizing both multi-model orchestration & efficiency without spending more.
Today we are launching Rinne, a harness which enables multi-model orchestration using whatever harnesses & APIs are available, fully open-source & local. Rinne also support loops while leveraging conductor, generator & evaluator architecture, all working together in the most efficient way possible.
Right now Rinne is a CLI based harness & v0.1.5 is fully stable. Rinne is written in Rust & thus providing a lightning fast user experience.
We are working towards a more robust & efficient User Experience with many more features coming soon.
Github - https://t.co/Hp94ntgxAN
Crates - https://t.co/d7nu7tRUPk
CC - @brooklyndotos @thedeepanshuweb
The best AI model is a trap. The teams pulling ahead orchestrates many.
We though why not build something which can use all your available harnesses & API, optimizing both multi-model orchestration & efficiency without spending more.
Today we are launching Rinne, a harness which enables multi-model orchestration using whatever harnesses & APIs are available, fully open-source & local. Rinne also support loops while leveraging conductor, generator & evaluator architecture, all working together in the most efficient way possible.
Right now Rinne is a CLI based harness & v0.1.5 is fully stable. Rinne is written in Rust & thus providing a lightning fast user experience.
We are working towards a more robust & efficient User Experience with many more features coming soon.
Github - https://t.co/Hp94ntgxAN
Crates - https://t.co/d7nu7tRUPk
CC - @brooklyndotos @thedeepanshuweb