Looking to connect: AI/quant builders, RAG engineers, ecom growth folks
I'm running three AI-first ventures simultaneously and I'm not looking for a co-founder in the traditional sense — I need sharp collaborators for specific problems.
Here's where I could use another brain:
-
SuperIntel / HedgeVision (autonomous AI for hedge funds)
Looking for: quant/algo trading background, RAG optimization experience, or anyone who's built at the intersection of LLMs + financial time-series. I have the architecture. Looking for someone to stress-test it, contribute, or go deep on a specific module. -
VIEngine (AI-first content + automation infra, beta in days)
Looking for: someone who's done B2B SaaS growth from 0→1000 users. Product is ready. Distribution is the next mountain. -
MangalMurti Jewellers (lab-grown diamond ecom, just launched)
Looking for: anyone with D2C ecom experience in India — jewellery, luxury, or adjacent. New to online, strong offline roots, genuinely good product.
What I bring back:
- Deep AI/data infra knowledge (RAG pipelines, LangGraph, FastAPI, vector DBs)
- Happy to advise, collab, or swap feedback on your project in return
- Long-term builder mindset, not a "ship and ghost" type
If any of these match what you're working on or what you know well — reply or DM. Let's find the overlap.
The SuperIntel / HedgeVision direction is the most interesting here — especially the intersection of LLM reasoning with financial time-series, which is still very under-explored compared to typical RAG use cases.
One thing I’ve been thinking about: most RAG pipelines break down in trading contexts because retrieval is static while markets are non-stationary. It feels like the real edge comes from combining time-aware retrieval (regime/context windows) with deterministic signals rather than relying purely on semantic similarity.
Also curious how you’re handling evaluation — backtesting LLM-assisted decisions is tricky since the model itself can introduce non-determinism. Are you anchoring outputs to structured signals before execution, or letting the model operate more freely in decision-making?
My background is more on the systems side (React/Node + backend infra), but I’ve been moving deeper into LLM workflows and data pipelines. Happy to stress-test parts of the architecture or go deep on a specific module if useful.
Hope we can connect and discuss it.
"Building autonomous RAG systems at the intersection of LLMs and financial time-series is high-level work—staying bootstrapped forces that close-to-user focus you mentioned.Nice idea, this could be a good way to test it. There’s a competition where you can submit it — entry is $19 and winner gets a Tokyo trip.Prize pool just opened at $0 so your odds are the best right now."