DP (Durgaprasad)

Founding member @vorfluxai, prev. founded @alpynelabs, studied @iitbombay

New account, old one got deleted.

[Vorflux @vorfluxai]

Obvious take: Testing becomes more acute when agents write code

Not-so-obvious take: Automatically run every service end to end with authentication, seed data, and a full integration test.

Most non-obvious take: Do it insanely subsidized by bringing your codex sub into vorflux

[Vorflux @vorfluxai]

New: @OpenRouter and @vercel AI Gateway as BYOK providers. Any model behind one harness.

[Vorflux @vorfluxai]

Table stakes, but ICYDK you can select any text in an agent response and ask a follow up with the quote attached to your message. Works on web and in the mobile app.

[Kush @kushbhuwalka]

@myprasanna yeah plan of plans is byfar the best part of vorflux. its probably wise to spend more time on this step than the actual implementation itsefl

[Prasanna S @myprasanna]

Yes this is why Vorflux employs a plan of plans and then summarizes to you aspects like architecture, data model and wireframes so you can error correct the places where RL hasn’t gone yet.

[Vorflux @vorfluxai]

THE DEATH OF SWITCHING COSTS

Is your software actually good, or are your customers just trapped? Hear from @eshamanideep on how he thinks about product quality in today's age

- Moving data between major platforms (like migrating from Salesforce to HubSpot) can now be done in as little as a single day.

- Legacy giants used to charge high renewal fees simply because switching away from them was too difficult. That advantage is completely gone.

- Competitors who prioritize rapid scaling and adding headcount over product quality end up losing customers.

- As switching costs drop to zero, the only way to retain and win customers is to focus ruthlessly on building the best possible product.

[DP (Durgaprasad) @dpm_2512]

@garg_tushar_ @rohanchdry @spc_india Interest if you think it’s a fit

[Vorflux @vorfluxai]

The longer the prompt, the worse the plan usually gets.

The model reads your repo faster than you can describe it. Tell it who the user is and what the outcome should be, then let explore sub-agents go find the rest.

[Vorflux @vorfluxai]

Machine setup in Vorflux is an agent session of its own. It scouts each repo for services, runtimes, ports and required env, writes a provisioning plan, submits it for review, then fans out one executor per repo.

[DP (Durgaprasad) @dpm_2512]

@sirsho29 https://t.co/BcYULWyGib

Can connect you the person who made ours

[Prasanna S @myprasanna]

Launching @vorfluxai : The autopilot for software engineering. I was prev co-founder / CTO of @Rippling ($10B) and #1 coder in India. Vorflux is my high octane Ferrari.

Every AI coding tool still makes you fly the plane. That's the copilot model: you stay in the seat, approving every turn. The models quietly got good enough to fly the whole route, but the tools never caught up.
So we built the autopilot.

@vorfluxai raised a $15M seed by @ycombinator @peakxvpartners @alliancedao @parkerconrad @jake_zeller @balajis @nivi @metakovan @lmrankhan @nikitabase @0xrwu @ayushjaiswal @mattshumer_ @eshamanideep @sreeramkannan @dvcoolster @nusimow @TeddySolomon11 @ashtoncofer @rvivek etc

Drop your biggest engineering bottleneck below. I'll reply with how I'd attack it with Vorflux, and hand you $200 in free credits to bang out your backlog.
Our full thesis 🧵👇

[DP (Durgaprasad) @dpm_2512]

@pettiboy_com Yes, welcome to the club. Also, carry your own bags to save environment 💰

[DP (Durgaprasad) @dpm_2512]

@SullyOmarr Hey, consider using @vorfluxai

It is a simple onboarding 1-click onboarding flow. It configures cloud machines by default and you can run all tasks without any restrictions - from debugging till testing.

If any help is needed, please do let me know.

[Shubham Saboo @Saboo_Shubham_]

This is the WAY.

Long running AI Agent harness where GPT 5.6 writes and Claude Opus 5 reviews.

Asked it to add Hermes Agent support to my memory bank plugin. Woke up to a tested PR.

[DP (Durgaprasad) @dpm_2512]

@kmish24 @myprasanna @Benioff @AnthropicAI @github @cognition @vercel @Dreamforce @SlackHQ @salesforce @agentforce @vorfluxai Hi @kmish24 do share here or over DM on how we can help more

[Prasanna S @myprasanna]

Live now! @vorfluxai office hours. Come join and ask your Qs.

https://t.co/SSFW7EftYv

[Vorflux @vorfluxai]

Building a Self-Driving Company Episode 5: Demo & Office Hours https://t.co/4vv1rG70f1

[Prasanna S @myprasanna]

Building a Self-Driving Company Episode 5: Demo & Office Hours https://t.co/DvF03sGIwr

[Chubby♨️ @kimmonismus]

1/ Reviewing agent-written code usually comes down to reading a diff and hoping.

That is the part of the workflow nobody solved. The models got very good at producing code, and the burden quietly moved to the person who has to decide whether it is safe to merge. You read it, you spot-check it, you run it locally if you have time, and then you approve on instinct.

@vorfluxai goes after that step directly. Every session spins up your whole stack on a real cloud machine. Repos, databases, CI/CD, feature flags, all of it. The actual environment your app runs in, with real services instead of mocks.

[DP (Durgaprasad) @dpm_2512]

@RavinBarthwal Hi Ravindra, can you share what you didn’t like about vorflux? Would love to learn your feedback

[DP (Durgaprasad) @dpm_2512]

@kimmonismus Thanks for featuring us