Sign in to view Shashwat’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Shashwat’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
India
Sign in to view Shashwat’s full profile
Shashwat can introduce you to 8 people at ConversAILabs
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
10K followers
500+ connections
Sign in to view Shashwat’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Shashwat
Shashwat can introduce you to 8 people at ConversAILabs
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Shashwat
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Shashwat’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
About
Welcome back
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
New to LinkedIn? Join now
Articles by Shashwat
-
AI Didn't Learn to Code. We Spent 50 Years Digitizing Knowledge
AI Didn't Learn to Code. We Spent 50 Years Digitizing Knowledge
Everyone says AI is replacing programmers because it learned to code. I think that's backwards.
10
1 Comment -
The AI Companies That Got Funded (And Why)Sep 21, 2025
The AI Companies That Got Funded (And Why)
I spent my weekend diving into AI companies that recently raised funding. What started as casual research turned into a…
29
1 Comment -
Unlocking the Potential of ChatGPT - A Game Changer for DevelopersJan 22, 2023
Unlocking the Potential of ChatGPT - A Game Changer for Developers
As a developer, you may have often come across tasks that seemed daunting and time-consuming. But what if I told you…
47
7 Comments
Activity
10K followers
-
Shashwat Gupta posted thisRs 19,000/month is all we need. That is roughly what I pay for ChatGPT. I already use it as my: → Chief of Staff → HR Manager → Senior Software Engineer → DevOps → Content Manager But the bigger change is not jobs. It is what we will expect from every software product. We are getting used to ChatGPT. I don't want to learn 20 screens or figure out workflows. I just tell it what I want. Soon we will expect every product to work like this: Understand → Use context → Act → Verify → Remember Companies can't just add an AI chatbot to their existing product. The product itself has to become AI native. And engineering changes with it. CRUD is increasingly solved. Engineers will need to understand agentic loops, tools, memory, evals and verification. We won't just build software people use. We will build software people delegate work to. Companies will have to become AI native because their users are becoming AI native first.
-
Shashwat Gupta posted thisIt was really sad to hear about the IIT Bombay student suicide case. Even 25 years back, IIT was such a big thing that somewhere you started believing: 𝗜𝗳 𝘆𝗼𝘂 𝗱𝗼𝗻'𝘁 𝗴𝗲𝘁 𝗶𝗻, 𝘆𝗼𝘂 𝗵𝗮𝘃𝗲 𝗮𝗹𝗿𝗲𝗮𝗱𝘆 𝗳𝗮𝗶𝗹𝗲𝗱 𝗶𝗻 𝗹𝗶𝗳𝗲. And once you are there, the stakes can feel even higher. A failed subject or repeated semester can start feeling much bigger than it really is. It took me years to understand that life doesn't depend only on which college you went to, but much more on what you learn, the skills you build, and what you are able to do with them. I sometimes feel we protect kids from failure too much. Maybe one way to change that is to let them experience failure early, when the stakes are still small. At 16–18, give them ₹50K and ask them to build a small business and try to make that ₹50K back. One condition: 𝘁𝗵𝗲𝘆 𝗵𝗮𝘃𝗲 𝘁𝗼 𝘀𝗲𝗹𝗹 𝗶𝘁 𝘁𝗵𝗲𝗺𝘀𝗲𝗹𝘃𝗲𝘀. Make 100 cold calls. Hear 99 NOs. They'll learn that rejection is normal. That you can work hard and still fail. That people won't always respond the way you expect. And most importantly: 𝗟𝗶𝗳𝗲 𝗶𝘀 𝗻𝗼𝘁 𝗹𝗶𝗻𝗲𝗮𝗿. Maybe that's the lesson we need kids to learn early. Things will go wrong. You will fail at some things. You change direction and try again. 𝗢𝗻𝗲 𝗳𝗮𝗶𝗹𝘂𝗿𝗲 𝗶𝘀 𝗻𝗲𝘃𝗲𝗿 𝘁𝗵𝗲 𝘄𝗵𝗼𝗹𝗲 𝘀𝘁𝗼𝗿𝘆.
-
Shashwat Gupta shared thisWhen did you last check Google Search Console, review your traffic, or publish a blog? As a founder, I kept postponing this. So at ConversAILabs, we started automating the routine with OpenAI ChatGPT/Codex. We connected Search Console, Google Analytics, BigQuery and GitHub to: → Track performance in a daily Google Sheet → Find searches with impressions but few clicks → Track ChatGPT referrals and landing pages → Publish three researched articles each week This week, we saw encouraging improvements: • Search impressions: +17.2% • Google clicks: +7.7% • Measured organic users: +34.9% Organic sessions fell 9.8%, but we reached more users, with fewer sessions per user. Overall, visibility and audience reach improved—though one week doesn’t establish a lasting trend. The useful part is the repeatable process: check the data, identify opportunities, publish, improve pages, and measure again.
-
Shashwat Gupta shared thisYour AI agents make mistakes. But do they learn? At ConversAILabs, we build conversational agents that improve through real conversations and your team’s feedback. Missed intent. Unclear answers. Poor follow-ups. Each one is an opportunity to improve—with your team reviewing and approving changes before they go live. Build better conversations into your business. Connect with us to explore what ConversAI Labs can do for your team. #ConversAILabs #ConversationalAI #AIAgents
-
Shashwat Gupta shared thisFrom a customer call to a confirmed booking—with the CRM updated and confirmation sent. A quick look at what we’re building at ConversAILabs. 100+ integrations to connect your calls with your business tools. What would you like to automate after a customer call? #ConversAILabs #VoiceAI
-
Shashwat Gupta shared thisYour support bot just promised a free replacement. Your policy never allowed it. Now your team has two problems: 1. Fixing the customer’s expectations 2. Figuring out why the bot said it. This is why we moved from one giant prompt to state-based conversation agents at ConversAILabs Each state has a defined job. Transitions make the conversation’s path visible. When something goes wrong, the team has a clearer starting point for debugging. Our principle is simple: Your bot should follow your knowledge, not invent answers. And your team should be able to trace what happened.
-
Shashwat Gupta shared thisThe call ends. The customer journey doesn’t. A good call means little if the next steps never happen—details aren’t sent, documents aren’t collected, and nobody follows up. That’s why voice AI in production needs a workflow beyond the call. At ConversAILabs , we’re building an AI operations employee that coordinates: • Voice calls and personalized WhatsApp follow-ups • Document collection and scheduled next steps • Conditional rules and human handoffs This short demo shows how those steps connect into one customer journey. Building Voice AI in Production — Beyond the Call. What still happens manually in your business after a customer call? #VoiceAI #WhatsAppAutomation #ConversAILabs
-
Shashwat Gupta shared thisVoice AI calls don’t follow a fixed path. Every caller can take a different direction—creating hundreds of possible scenarios and unexpected edge cases. That makes safely editing and testing voice AI workflows extremely difficult. At ConversAILabs , we’ve built a state-of-the-art agentic workflow editor to solve this. Simply describe the change you need. The editor updates your agent, validates every affected conversation path and automatically generates relevant tests. Build faster. Test more scenarios. Ship reliable voice agents with confidence. Here’s a 30-second look. Note: The presenter is AI-generated; the product footage is from our actual workflow editor. #VoiceAI #AgenticAI #ConversAILabs
-
Shashwat Gupta reposted thisShashwat Gupta reposted thisWhat happens when a customer misses your AI’s outbound call—and calls the same number back? With ConversAI Labs, the AI answers the callback and continues the relevant conversation. Outbound + inbound. Same AI agent. No lost conversation. conversailabs.com #VoiceAI #ConversationalAI #SaaS
-
Shashwat Gupta liked thisShashwat Gupta liked thisWe're rebranding to Cashew Labs and been working on something very cool the past few weeks. Although the ideas have been cooking for me since I first got into computers Excited to share more this week!
-
Shashwat Gupta liked thisShashwat Gupta liked thisWe’re growing our specialist network at Broadvale AI. Some projects need deep expertise in a specific area. We’re looking to work with independent AI engineers and specialists who care about building reliable, production-ready systems. Your expertise may be in LLM fine-tuning, RAG, agentic automation, evaluation, AI product engineering, or another specialised area. When the right project comes along, you join the team as an independent consultant. We handle the client relationship and project coordination. You bring your expertise and work closely with us to deliver. If this sounds interesting, send me a DM with your portfolio, area of expertise, and availability. Or leave a comment below and I’ll reach out. Looking forward to build some awesome AI Systems together.
-
Shashwat Gupta liked thisShashwat Gupta liked thisYesterday, someone from Zambia reached out to Bloom Femtech for a consultation. 🇿🇲 It was our first consultation from outside India, and honestly, it felt pretty special. You build for the people around you, so having someone thousands of kilometres away find the platform and book a consultation was a moment I really enjoyed. #Bloom #WomensHealth #FemTech #DigitalHealth #BuildInPublic
-
Shashwat Gupta liked thisRs 19,000/month is all we need. That is roughly what I pay for ChatGPT. I already use it as my: → Chief of Staff → HR Manager → Senior Software Engineer → DevOps → Content Manager But the bigger change is not jobs. It is what we will expect from every software product. We are getting used to ChatGPT. I don't want to learn 20 screens or figure out workflows. I just tell it what I want. Soon we will expect every product to work like this: Understand → Use context → Act → Verify → Remember Companies can't just add an AI chatbot to their existing product. The product itself has to become AI native. And engineering changes with it. CRUD is increasingly solved. Engineers will need to understand agentic loops, tools, memory, evals and verification. We won't just build software people use. We will build software people delegate work to. Companies will have to become AI native because their users are becoming AI native first.
-
Shashwat Gupta liked thisShashwat Gupta liked thisDear PMs, how would you interpret this cohort retention data? We’ve been tracking Bloom Femtech's retention cohort-by-cohort as acquisition has scaled significantly. Would love to hear how experienced PMs would read these numbers? Open to perspectives, benchmarks, and especially what you would investigate next.
-
Shashwat Gupta liked thisShashwat Gupta liked thisEVERQ just made me look at corporate life a little differently. 😂 We’ve somehow reached a point where looking exhausted is treated like proof that you’re working hard. Dark circles? “Busy.” Tired face? “Hardworking.” Looking fresh? “Do you even have enough work?” �� That’s exactly what I liked about EVERQ ad it takes this very relatable workplace expectation and makes fun of it. EVERQ is built around skincare that fits into everyday routines, because taking care of your skin shouldn’t feel like another task on your corporate to-do list. And honestly, I love the idea that looking fresh doesn’t mean you’re working less. Your work should speak for itself. Your dark circles don’t need to. 😂 #SoDermatic #EVERQOrignals #DermaMeetsDerma SKINQ
-
Shashwat Gupta liked thisShashwat Gupta liked thisI'm 46. Here's my tech life in one scroll. Grew up with no computer at home. Phone calls meant the STD booth after 9 PM because rates dropped. Then dial-up. That screeching modem, someone at home picking up the landline mid-download. One banner downloading for four minutes, top to bottom, like a curtain rising. Then Orkut, testimonials and scraps. Then Gmail invites, treated like club memberships. Then e-commerce. Cash on delivery, because nobody trusted a website with their card. I know, because I was running one of those websites. Then UPI. Now the cab driver has a QR code and I have no change. And now an AI concierge books my dinner and checks me into flights while I sleep. Every five years I thought I'd seen the big shift. Every five years I was wrong. Somewhere between the modem and the agents, I stopped being surprised and started feeling lucky. Most generations get one wave. Ours got all of them.
-
Shashwat Gupta liked thisShashwat Gupta liked thisReticle 𝘀𝘁𝗮𝗿𝘁𝗲𝗱 𝘄𝗶𝘁𝗵 𝗮 𝗖𝗹𝗮𝘂𝗱𝗲 𝗹𝗶𝗺𝗶𝘁 𝗮𝗯𝗼𝘂𝘁 𝘁𝗼 𝗲𝘅𝗽𝗶𝗿𝗲 One night, Divyanshu had 25% of his weekly limit left, and it was about to expire so he did the obvious thing he gave Claude one massive prompt for a side project and went to sleep next morning, Claude said it was done and everything is working 𝗘𝗫𝗖𝗘𝗣𝗧 𝗜𝗧 𝗪𝗔𝗦𝗡'𝗧 buttons didn’t work parts of the app were running on mock data things that looked finished were completely broken underneath... and Claude had absolutely no idea that was the part that stuck with us AI had become incredibly good at writing code But after writing it, the same loop kept happening: Agent writes -> agent says “done” -> human checks -> human finds the bug -> human explains it back to the agent we were becoming our coding agent’s feedback loop the next week, we started building Reticle The idea was simple: Reticle gives coding agents a way to prove their work actually works it looks at the running application, checks what actually happened, and gives that evidence back to the agent The agent still writes the code Reticle makes it prove it Today, we’re putting Reticle out there properly 𝗕𝗲𝗶𝗻𝗴 𝘂𝘀𝗲𝗱 𝗯𝘆 𝟭𝟬𝟬𝟬𝘀 𝗼𝗳 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿𝘀 𝗴𝗹𝗼𝗯𝗮𝗹𝗹𝘆 This video is the story of why we built it, how it works, and where we think AI-assisted software development is heading. Try RETICLE!!
Experience & Education
-
ConversAILabs
**********
-
*******
***** ********** *******
-
***********
***** ********** *******
-
****** ********* ** *********** *******
****** ******** ******* undefined
-
-
******** *********
-
View Shashwat’s full experience
See their title, tenure and more.
Welcome back
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
New to LinkedIn? Join now
or
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View Shashwat’s full profile
-
See who you know in common
-
Get introduced
-
Contact Shashwat directly
Other similar profiles
Explore more posts
-
Palash S.
Ascend Ventures • 1K followers
Traditional, non-sales Indian tech founders who run IT consulting or custom software development agencies, typically: Started from coding background (Java, .NET, PHP, MERN, etc.) Rely heavily on referrals + old clients No formal sales system Team size: 5–50 engineers Revenue: ₹30 lakh – ₹5 crore/year Based in Tier 1 & Tier 2 cities (Indore, Pune, Ahmedabad, Bangalore, Hyderabad, Delhi, Mumbai) 5–12+ years in business Strong delivery capability but weak branding + weak founder-led sales --- Core Pain Point - Flat or Declining Revenue 1. Pipeline Dying Slowly Referrals slowing down Old clients reducing spend No predictable new leads Dependency on luck and old relationships 2. No Personal Brand = No Inbound Founders don’t show their face online Bare minimum LinkedIn activity Zero thought leadership Prospects can’t differentiate them from 5000 other IT companies 3. Founder is Stuck in Operations 90% time goes in delivery, firefighting, team management 0% time goes into marketing, sales, building authority Result: Revenue stagnates even if delivery is excellent 4. No Sales System, Just Hope No positioning No niche No offer clarity No content system No outreach system No lead-nurturing No inbound funnel 5. Projects Coming but Profits Falling New projects are low-ticket Clients bargain Too much dependency on Upwork/Agencies Margins dropping. 6. Competition Growing Faster Younger founders using LinkedIn + personal branding
6
2 Comments -
Dhruvin Desai
ProductSquads • 5K followers
Xiomi in English vs Xiomi in Gujarati - how we could have saved 75 Billion parameters. Chain of thoughts - that lead from a Signage Board of 'Xiomi' in Gujarati in The Karnavati City (aka Ahmedabad). Typically Xiomi has spend good amount of PR campaign in making its pronunciation easy or actually make it right, as its not very intuitive for someone to read 'Sh' sound from English 'X'. But in Gujarati and many Indian languages the way you pronounce and the way you write is exactly same. Something that 'Xiomi' spend good amount of PR and leadership campaign for pronunciation to be right was effortlessly delivered on Shop Signage Board. Even for somebody who is introduced to brand for the first time. It's an Global brand that needs acceptance in local market, of course the best way to get local acceptance is to get messaging in local language, in this case xiomi is getting back for free that is PR campaign cannot delivery by just use of structured language. Then i thought how much of today's LLMs parameters out of trillions are actually trying to solve for 'English' as a language that does not have intuitive or programmable logic to generate right pronunciations which are right for each region (Remember Xiomi's is chinese ) and context. Not only pronunciation that it generates but also when user from different Accent tries to provide an input. Would we have saved significant parameters of large model ? "Research into specialized models (like HindiLLM and Sarvam) suggests that by using a language where syntax is deterministic (like Sanskrit or high Hindi), you could likely reduce the parameter count by 15–20% without losing reasoning capability. The model wouldn't need as many "attention heads" dedicated solely to resolving linguistic ambiguity" " If we had a "Native Hindi" model with an optimized tokenizer, a sentence that takes 10 tokens in English might only take 6–7 in Hindi while carrying more logical weight. This would allow a model with 30% less memory (smaller context window) to "remember" the same amount of information." Globalization created unstructured language like 'English' as default, if not for English NLP accuracy and reasoning could have arrived much earlier. "In English, a sentence like "I saw the man with the telescope" is structurally ambiguous (Who has the telescope?). Researchers spent nearly 30 years (the "AI Winter" period) just building complex rule-bases to solve these edge cases. We might have bypassed the "Symbolic AI" failures of the 70s and 80s, potentially reaching the Neural Revolution (Deep Learning) a decade earlier because the "data" would have been naturally machine-readable" Globalization's this aspect in terms of added cost to research and humanities progress is an intriguing thought, globalization should adapt to local rather changing the local for its own success.!!!!
9
-
Sathyanarayana RB
Clientseva LLP • 488 followers
🏭 THE MANUFACTURER THAT THOUGHT IT NEEDED MORE CUSTOMERS… …but its data told the CXO something completely different. Imagine a medium-sized Tier-2 engineering manufacturer with annual revenue of ₹120 Cr. Growth had slowed: Year 1: ₹120 Cr → Year 2: ₹132 Cr → Year 3: ₹138 Cr The management believed: «“We need more customers.”» Before increasing the sales budget, they decided to ask: “What is our existing data telling us?” They analysed 5 years of: 📊 Customer & product sales 💰 Pricing & profitability 🏭 Production & capacity 🔧 Quality & rejection 📦 Inventory 🚚 Delivery performance 📈 Monthly demand 🤝 Customer behaviour And the insights were surprising. 🔹 High-revenue customers were not always the most profitable. 🔹 Some high-volume products were consuming capacity but generating thin margins. 🔹 Certain regions had strong demand but inadequate sales focus. 🔹 Price sensitivity varied significantly across customer segments. 🔹 Quality and delivery issues were affecting repeat business. 🔹 Some customers showed declining purchase patterns before the sales team noticed. 🔬 Then XLSTAT changed the conversation. Using statistical analysis, regression, forecasting, segmentation and optimisation, management could move from: “What happened?” to: “Why did it happen?” “What will happen next?” “What should we change?” Pricing analytics helped identify opportunities to improve price and contribution. Forecasting helped improve demand and production planning. Customer and product analytics helped identify where sales effort and capacity could generate better returns. Quality analytics helped identify patterns behind rejection and rework. 📈 THE RESULT? Not simply “sell more.” The strategy became: SELL SMARTER. PRICE SMARTER. PRODUCE SMARTER. FORECAST SMARTER. Illustrative revenue journey: ₹120 Cr → ₹132 Cr → ₹138 Cr → ₹158 Cr → ₹192 Cr → ₹225 Cr → ₹260 Cr Illustrative scenario only—not a claimed customer result. Actual outcomes depend on data, market conditions and execution. The important lesson: Your next revenue opportunity may already be hiding inside your existing data. Your ERP has the data. Your sales team has the data. Your factory has the data The question is: Are you only recording what happened—or using data to decide what happens next? That is where Data Analytics + XLSTAT can become a powerful CXO decision-making tool. At ClientSeva, we help manufacturing organisations explore analytics across: Revenue | Pricing | Customers | Products | Forecasting | Quality | Inventory | Capacity | Profitability | Risk 🎯 CXOs / CEOs / CFOs / COOs / Plant Heads / Business Heads If you believe your company's data has untapped potential: DM “XLSTAT” Let's turn: DATA → INSIGHT → DECISION → ACTION → GROWTH 📈 #Manufacturing #Tier2Manufacturing #ManufacturingIndia #DataAnalytics #XLSTAT #RevenueGrowth #PredictiveAnalytics #ManufacturingExcellence #CXO #BusinessAnalytics #DigitalTransformation #Profitability #ClientSeva
1
-
Rajeev Mamidanna Patro
punnedit • 8K followers
Most IT channel & SaaS founders suffer from the same problem: Fear of missing out (FOMO). - Not the lack of capability - Not the lack of OEM relationships - Not the lack of services But for lack of clarity. “If I do 20 things, why show only 1 or 2?” “If I leave something out, what if I lose business?” And that is where your brand takes a beating. - You try to squeeze every OEM logo into one card. - You try to communicate every offering in one ad. - You try to be remembered for everything. And when you say everything, nothing sticks. Strong positioning is subtraction. Not addition. Your market remembers you for the one thing you lead with. Not the 4 page catalog you carry. Customers don't have time to meet you and 100 others like you to understand what you do better. Growth is simple: Choose what you can solve best and stand for it. Say only that. Your communication needs to be in their radar already for them to call you. Let the rest follow after people start trusting you. FOMO complicates. Clarity wins. ---- Rajeev Mamidanna Fixing what most tech founders miss out, Brand Strategy, Marketing Systems and Unified Messaging in 90 days and helping you with continuous Marketing
9
6 Comments -
Akhil Raj Kothoor
axldrone • 643 followers
"ActionFi as an action thriller" Would you be surprised if we told you that there is a company in Kozhikode that manufactures drones for the Indian Army? Kozhikode is the main hub of a company that manufactures drones for border surveillance for the Indian Army. actionfi is an IT company that has proven its excellence in drone manufacturing in addition to software manufacturing. Starting from Kozhikode and with offices and staff in the US, Actionfi has grown from just 5 people to an IT company with over 150 employees and a value of over 180 crores. The founder is Joe Kuncheria, a native of Palakkad, and his relatives Robin and Joseph are the managing partners. Joe, who has been working in the US for 25 years, was the vice president of an IT company. He met Rajeesh, a native of Kunnamangalam, who was starting a company in India for India operations. Rajeesh, who wanted to settle in his own country, came up with the idea of starting the company in Kozhikode. Initially, Actionfi started in a small space in Hilite Business Park. There were five people including Rajeesh, the general manager. Initially, it was developed to help large companies manage their finances. With the success of the automation software RunMyBot.The business began to grow. Today, 180 users in more than fifty countries use the RunMyBot software. A company called Axldrone was started in 2020. Initially, the work involved importing drone frames from other countries. The software required for the drones was developed by the Kozhikode campus. Team, realizing that imported drone frames were not enough for their needs, prepared its own manufacturing plan and opened an office in Kakkanad, Kochi. Electronics engineers and technicians were assigned here to design the frames for the drones and manufacture other electronic devices. Later, Joe Kuncheria also started a drone manufacturing unit on his family land in Kannambara, Palakkad. The Indian Army, recognizing the quality of Axldrone's products, requested it to manufacture drones for the Army. The drones, including many technologies such as thermal imaging and night vision, were manufactured for the Army. AxlDrone is preparing to manufacture more drones for the Army. Axldrone has brought Drones and Baseboards to the market for students to do projects. Plans are also underway to build various types of drones, including sound drones (a method of flying drones in groups) for drone shows. ActionFi and its sister company Axldrone are preparing to manufacture drones for various needs for Kerala government departments and anti-jamming software and equipment for the security of oil refineries and other strategic institutions, as per the strategic partnership signed with Keltron." #30Drones30Days
51
8 Comments -
Akshay Sole
Supeek Agro Marts • 166 followers
Tips for small businesses in India 1. Validate Your Idea Before Investing • Talk to potential customers • Run small paid ads to test interest • Sell a minimum viable product (MVP) first Goal: Confirm real demand before spending big. 📊 2. Know Your Target Customer Deeply Ask: • Who are they? • Where do they buy now? • What problem are you solving? • What price will they pay? Customer clarity = easier marketing + higher conversions. 💡 3. Start Lean, Spend Smart Keep early costs low by: • Using shared office space / working from home • Outsourcing tasks instead of hiring full-time • Using free or inexpensive tools (Google Workspace, Canva, Zoho) 📌 4. Register Properly (Legal & Tax) Common registrations for Indian small business: • Sole Proprietorship / Partnership • GST registration (if turnover > threshold or selling interstate/online) • Udyam Registration (MSME) – government benefits, subsidies 👉 Legal compliance builds trust and avoids penalties. 🚀 5. Build an Online Presence (Most Critical) • Website (even a basic one) • Google Business Profile • Social media channels relevant to your audience Listing your business online increases visibility and trust. 📣 6. Create a Simple Digital Marketing Plan Channels that work well in India: 📌 Social Media • Instagram Reels • YouTube Shorts • Facebook posts & groups 📌 Search Ads & Social Ads • Google Ads • Meta Ads (Instagram & Facebook) Start with a small budget to test what works. 🤝 7. Deliver Great Customer Experience Delighting customers builds: • Referrals • Repeat business • Positive reviews Always follow up after a sale. 💰 8. Price Your Product/Service Right Consider: • Cost of making + delivering • Competitor prices • What customers are willing to pay Avoid pricing too low — sometimes higher pricing = premium perception. 🧠 9. Track Metrics (Numbers Matter) Track these regularly: • Sales revenue • Customer acquisition cost • Profit margin • Repeat purchase rate Data lets you improve decisions fast. 🤝 10. Collaborate & Network Join: • Local business groups • Industry associations • Online communities (WhatsApp, LinkedIn) Collaborations can open new customers and partnerships. 📦 11. Get Online Sales Channels Right For product businesses: • Sell on your own website (higher margins) • List on marketplaces: • Amazon • Flipkart • Meesho / other reseller platforms Use COD and easy returns (Indian consumer expectation). 📊 12. Focus on Customer Retention 80% of revenue comes from 20% of customers. Tools to keep customers: • WhatsApp broadcast lists • Email newsletters • Loyalty programs 📈 13. Scale When Ready After stable growth: • Hire freelancers or part-time help • Automate repetitive tasks
6
-
Indrajit S.
CellStrat • 7K followers
Indian clinics are bleeding ₹50,000+ every month. Not from bad doctors. Not from poor care. From a ringing phone — and no one to answer it. When Vivek and I spent weeks inside local clinics across Bangalore, we kept tripping over the same silent revenue leak: The Missed Call. A patient dials during peak hours. The line is busy. They don't wait. They open Google, call the next clinic, and book there. Your receptionist was on lunch. You just lost a patient — and probably ₹2,000–5,000 in lifetime value. Multiply that across a day. A week. A month. The math is brutal. So we stopped sleeping, drank a lot of chai ☕, and built the solution. 😉 Today, we're officially launching cellAssist.ai — a vernacular Voice AI front-desk, built specifically for the realities of Indian healthcare. Not a clunky chatbot. Not another HMS nobody uses. A Voice AI that: → Answers every call, 24/7 → Speaks 20+ Indian languages → Books the appointment straight to the clinic's WhatsApp in under 60 seconds → Requires zero training for your existing staff None of this would exist without our engineering and product pod, who cracked one of the complex problems in this space: running low-latency LLMs over noisy Indian cellular networks. Massive respect also to our GTM team already out in the field — we're onboarding our first cohort of 20 clinics across India for a 2-week live performance pilot right now. If you run a clinic, or know a doctor whose front desk is overwhelmed — DM me, or visit cellAssist.ai We can have your AI receptionist answering phones by tomorrow morning. #HealthTech #IndiaHealthcare #VoiceAI #AIStartup #DigitalHealth #cellassist #AI #ClinicManagement #HealthcareInnovation #MadeInIndia #StartupIndia #BangaloreStartups #PatientExperience #AIReceptionist #HealthcareAI #IndianStartups #FutureOfHealth CellStrat Healthcare
135
41 Comments -
Sunny Narang
Center for Embodied Knowledge • 5K followers
#StartupMafia 3.0… and the Coming Startup Mafia 4.0 An interesting story in the The Economic Times today, by Disha Acharya & Puran Choudhary, highlights how companies like Razorpay, Cred, Meesho and PhonePe have become “founder factories,” spawning 200+ new startups. This phenomenon—often called the Startup Mafia effect—is actually a sign that an innovation ecosystem has matured. Silicon Valley saw it first with the PayPal Mafia. India is now seeing the same pattern. Startup Mafia 1.0 (2007–2016) The first wave came from pioneers like Flipkart and Paytm. This era built India’s consumer internet economy. Focus sectors included: • e-commerce • online marketplaces • payments • consumer apps This phase proved that large digital businesses could be built in India. Startup Mafia 2.0 (2014–2021) The second wave rode the mobile internet explosion after Jio. Companies like: • Ola • Oyo • Swiggy • Zomato trained a generation of founders who understood scaling operations, logistics and hypergrowth. India’s startup ecosystem moved from experimentation to unicorn-scale companies. Startup Mafia 3.0 (2018–Present) Today’s wave is driven by companies like: • Razorpay • Cred • Meesho • PhonePe The backbone of this phase is India’s digital public infrastructure: UPI Aadhaar cheap cloud infrastructure developer ecosystems Many new startups emerging from these companies focus on: • fintech infrastructure • AI-native software • cross-border payments • SaaS tools. Startup Mafia 4.0 (Emerging) But the next wave may look completely different. Instead of consumer apps, it will likely come from deep-tech ecosystems. Places like: • IIT Madras Research Park • semiconductor clusters in Bengaluru and Hyderabad • robotics and EV startups • industrial analytics companies. Here we see companies like: • Ather Energy • AgniKul Cosmos • Detect Technologies • Gyan Data training engineers who will eventually become founders themselves. The earlier waves digitised commerce and payments. The next wave may digitise machines and infrastructure. Instead of apps connecting people… we will see platforms connecting: • factories • vehicles • industrial machines • supply chains. This is where mechatronics + IoT + AI come together. If sensor data, embedded electronics and cloud analytics are layered onto these machines, they can become connected intelligent assets. Predictive maintenance, uptime guarantees and performance optimisation become possible—even for affordable machines. Industry 4.0 for the Real Economy Germany built Industry 4.0 for high-end factories. India may build Industry 4.0 for millions of machines used across the Global South. And that next generation of founders—Startup Mafia 4.0—may emerge not from consumer apps, but from mechatronics, IoT and industrial intelligence ecosystems. The next startup revolution may not just connect people. It may connect the machines that power the real economy.
5
-
Prasanna K Ram
ChatOps.health • 7K followers
Why India doesn't need HIPAA (and why that's actually fine) HIPAA vs DPDP — think vehicle design rules vs traffic rules. HIPAA is like vehicle design standards. It specifies how a vehicle must be built, tested, and certified. Detailed. Prescriptive. Healthcare-specific. DPDP is more like traffic rules. It doesn't redesign the vehicle. It ensures drivers stay in their lane, speed is controlled, and there's a license to revoke when someone crashes. India never had healthcare-specific vehicle design rules for data. But most accidents weren't happening because vehicles were poorly designed. They were happening because everyone was driving their own way: Patient data floating in WhatsApp groups No audit trails Oversharing by default Zero accountability In my experience across healthcare tech, the problem isn't lack of advanced safeguards. It's lack of discipline. DPDP fixes discipline. HIPAA makes sense in a mature, tightly regulated ecosystem. In India today, it would often be overkill. DPDP is: Lighter Pragmatic Enforceable May not be perfect. But it's just enough — and necessary — for India right now.
24
17 Comments -
Mrinal Tyagi
HeyDoc AI • 5K followers
HeyDoc AI is already making waves across healthcare and digital health while we launch India's first #Agentic AI driven Multi specialty Hospitals with Clearmedi Healthcare (A Morgan Stanley company). HeyDoc AI #AI #HeyDocAI #AIinHealthcare #Trust #IndiaAI #GoogleForStartups #ClearMedi #ArtificialIntelligence #Agents Y Combinator #US #digitalHealth
15
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore MoreOthers named Shashwat Gupta in India
-
Shashwat Gupta
Bengaluru -
Shashwat Gupta
Greater Allahabad Area -
Shashwat Gupta
Mumbai -
Shashwat Gupta
Ghaziabad -
Shashwat Gupta
New Delhi
413 others named Shashwat Gupta in India are on LinkedIn
See others named Shashwat Gupta