Hey everyone! It's been a while since my last update here in the community. Three months ago I shared that I had closed my first audit and kicked off a new project — and since then, a lot has happened! Over these past few months I've been heads down developing it, and we've now officially set a go-live date and a delivery milestone. It's been an incredible journey and I'm so proud of how far this has come. Can't wait to close even more clients and keep the momentum going! Thank you for your attention Kunal

Something I've Realized While Learning AI The more I learn about AI, the more I think beginners focus too much on tools. I know I did. At first, I wanted to learn: ChatGPT AI agents Automations Every new AI tool But recently I've been noticing something: The people getting results don't seem obsessed with tools. They're obsessed with solving problems. For example: Instead of asking: "What AI tool should I learn next?" They ask: "What problem can I solve?" That shift feels small, but I think it changes everything. I'm still learning myself, but this has been one of the biggest lessons I've picked up so far. Curious what everyone thinks: What's been the most valuable lesson you've learned while building with AI?

Interesting constraint: most of their inbound is in Kannada. Standard English-language voice agents are useless here — the moment a customer hears an accent that doesn't match their context, they hang up. Two weeks of testing went into getting the Kannada TTS to not sound robotic. Switched from ElevenLabs to Sarvam. Iterated on the prompt structure 9 times. Added Kannada filler words ("hmm", "haan") to make pauses feel natural. The technical work nobody talks about: voice agents in vernacular languages are 10x harder than English-only deployments. Every word has to feel native. One robotic syllable and the call is dead. But the market opportunity is also 10x bigger. 600M+ people in India transact in vernacular languages. Almost nobody is building serious voice infrastructure for them. Reminder for anyone building: the languages, markets, and demographics that get ignored by global products are usually the highest-margin opportunities for local operators.

Hey everyone, I’m getting close to delivering my first project, but I’m still struggling with the prompt and I can’t seem to get to a final version that works properly. The frustrating part is that I had a prompt that was working quite well at the beginning. Then I started testing it with other people, took notes on everything that worked and everything that didn’t, and tried to improve it based on those issues. But once I started adding fixes and rewriting parts of the prompt, I kind of lost the balance. Now the agent doesn’t respond as naturally or as reliably as before, and I feel like I’m going in circles trying to fix one problem while creating another. Has anyone dealt with this before? I’d really appreciate any advice on how to approach this properly, especially when it comes to simplifying the prompt, avoiding too many rules, and getting the agent back to responding in a clean and natural way. Thanks in advance. Any help would be really appreciated.

Hey everyone! I want to show you an updated vesion of my previous real-estate automation. Here's exactly what happens when a lead submits a form: 1️⃣ Their details are instantly logged in Airtable 2️⃣ They receive a personalised welcome email within seconds 3️⃣ The agent gets an alert with the full lead details 4️⃣ A 48-hour countdown timer starts automatically If the agent hasn't contacted them after 48 hours: 5️⃣ A follow-up email sends automatically 6️⃣ If still no reply — repeats up to 3 times over the next few days 7️⃣ After 3 attempts with no reply — marked as Lost and stops forever The agent can manually stop the email sequence by checking "Stop Chasing" field in the airtable. I would appreciate any feedback about it!