Omkar Bansod

Software Developer | Building Ai Agents

[Omkar Bansod @_Omkar_Bansod]

@siddhantmohan1 W product, bigger W distribution 💥

[Omkar Bansod @_Omkar_Bansod]

@RahulKumaran4 @TryBrowzer @ProductHunt Congratulations 🎉 Yc soon? 👀

[Omkar Bansod @_Omkar_Bansod]

@harrychow_ @getsolari Your hiring approach is really cool! you get real users, real reach and some crazy builders.
But it feels a little disorganized. Maybe open a Google Form, bring selected people into Discord, give them a few use cases and a fixed build window. Would make evaluation easier too.

[Omkar Bansod @_Omkar_Bansod]

@YeshodharaB @comeonaditya Dmed you

[Omkar Bansod @_Omkar_Bansod]

@abhisheknaironx Us moment bhai 😭

[Omkar Bansod @_Omkar_Bansod]

@the_versecafe @gitcafe @ycombinator Congrats on the launch 🚀

[Omkar Bansod @_Omkar_Bansod]

@RahulKumaran4 Most useful thing I’ve seen on X since @thsottiaux reset our limits this week 🤌💥

You earned a follower.

[Omkar Bansod @_Omkar_Bansod]

@devxpy Congratulations 🎉

[Omkar Bansod @_Omkar_Bansod]

@Amboras_inc Applied.

Had already spent some time building around Amboras’ problem space, so resurfacing StorePilot here.
https://t.co/trddlQjzFT

[Omkar Bansod @_Omkar_Bansod]

Saw @Amboras_inc is hiring backend/full-stack engineers, so I spent some time understanding what they’re building.

What stood out to me was the part after a store goes live, using real shopper behavior to keep improving the storefront.

So I built StorePilot, a focused proof-of-work around that loop:

product page → funnel events → revenue analytics → experiment lab → Bayesian confidence → promote winner

Core loop is deterministic: events, attribution, analytics, confidence, and promotion. AI can assist variant copy, but the system owns the decision.

Would love feedback from the Amboras team.
@imad_mkdm @Amin_mokdm @Vaibhav30665241

[Omkar Bansod @_Omkar_Bansod]

@TonyXavier_ Dmed you 🤝

[Omkar Bansod @_Omkar_Bansod]

Been exploring how embedding spaces can become something people can actually navigate, not just vectors sitting in a DB.

Built Latent Atlas : real @metmuseum artworks �� CLIP embeddings → UMAP + HDBSCAN → interactive visual exploration.

This started after seeing some of the problems @GooeyAI is exploring around cultural datasets.

oss : https://t.co/ExpsM9bpTs

[Omkar Bansod @_Omkar_Bansod]

What if the conversation itself was your document editor?

So I built Draftcord: a document editing agent inside Discord, powered by @superdocsapp.

Upload a DOCX → describe changes in plain English → review the exact edits → approve/reject → export the updated DOCX or PDF.

No jumping between chat and an editor. The conversation becomes the document workflow.

SuperDocs puts it perfectly: “Your AI can write. It’s time it could edit.”

Really impressed by what @OmkaraKadam is building at SuperDocs.

[Omkar Bansod @_Omkar_Bansod]

@tonychenAI sounds like the bug loop needs another engineer 👀 sent you a DM a while back.

[Omkar Bansod @_Omkar_Bansod]

@myprasanna @vorfluxai @Rippling Looks exciting! Congratulations on the launch 🚀

[Omkar Bansod @_Omkar_Bansod]

@smthomas3 @mastra Dmed you 🤝

[Omkar Bansod @_Omkar_Bansod]

@kushagrchitkar Docsflix and chill 😂

[Omkar Bansod @_Omkar_Bansod]

@rohansharma0509 Congrats! @usealmanac is solving a real pain.

AI coding agents can read code, but they still miss the “why” behind decisions, gotchas, and old tradeoffs.

Will try it out.

[Omkar Bansod @_Omkar_Bansod]

@kushal124 Hey @kushal124 Here is a project i recently worked on sounds similar with your product Idea : https://t.co/OMLV2PY8F5

Would love your feedback!

[Omkar Bansod @_Omkar_Bansod]

Saw @Amboras_inc is hiring backend/full-stack engineers, so I spent some time understanding what they’re building.

What stood out to me was the part after a store goes live, using real shopper behavior to keep improving the storefront.

So I built StorePilot, a focused proof-of-work around that loop:

product page → funnel events → revenue analytics → experiment lab → Bayesian confidence → promote winner

Core loop is deterministic: events, attribution, analytics, confidence, and promotion. AI can assist variant copy, but the system owns the decision.

Would love feedback from the Amboras team.
@imad_mkdm @Amin_mokdm @Vaibhav30665241

[Omkar Bansod @_Omkar_Bansod]

@Amboras_inc Built a small proof-of-work after seeing this opening.
Shared the demo here: https://t.co/trddlQjzFT
Would love feedback.

[Omkar Bansod @_Omkar_Bansod]

Saw @Amboras_inc is hiring backend/full-stack engineers, so I spent some time understanding what they’re building.

What stood out to me was the part after a store goes live, using real shopper behavior to keep improving the storefront.

So I built StorePilot, a focused proof-of-work around that loop:

product page → funnel events → revenue analytics → experiment lab → Bayesian confidence → promote winner

Core loop is deterministic: events, attribution, analytics, confidence, and promotion. AI can assist variant copy, but the system owns the decision.

Would love feedback from the Amboras team.
@imad_mkdm @Amin_mokdm @Vaibhav30665241

[Omkar Bansod @_Omkar_Bansod]

Saw @Amboras_inc is hiring backend/full-stack engineers, so I spent some time understanding what they’re building.

What stood out to me was the part after a store goes live, using real shopper behavior to keep improving the storefront.

So I built StorePilot, a focused proof-of-work around that loop:

product page → funnel events → revenue analytics → experiment lab → Bayesian confidence → promote winner

Core loop is deterministic: events, attribution, analytics, confidence, and promotion. AI can assist variant copy, but the system owns the decision.

Would love feedback from the Amboras team.
@imad_mkdm @Amin_mokdm @Vaibhav30665241