Sudipta Biswas
San Francisco, California, United States
16K followers
500+ connections
About
Doing sales is harder than ever. There are too many people selling to too many potential…
Activity
16K followers
-
Sudipta Biswas shared thisRIP stale database! Your next customer isn't sitting in a database from 2022. They're out there right now - Agent Jesse finds them live! Apollo: stored database scraped years ago Agent Jesse: searched now from live internet!. Stop searching snapshots! Try Agent Jesse - https://agentjesse.ai/ Agent Jesse
-
Sudipta Biswas shared this"Better than LinkedIn's own native tools." That's how Oliver described Jesse after using it to build prospect lists for his coaching business. Finding the right academic decision-makers used to mean hours of searching university websites, LinkedIn, and Google. Then he tried Jesse. The search results were exactly what he needed, with contact and LinkedIn data that was close to 100% accurate in his experience! "As a tool for screening and vetting prospects on LinkedIn, it's simply better than anything else I've used - including LinkedIn's own native tools." From there, he exported everything into a spreadsheet and handled outreach his own way. Feedback like Oliver's helps shape where Agent Jesse goes next. If finding the right people matters more than finding more people, give Jesse a try. https://agentjesse.ai/ Agent Jesse
-
Sudipta Biswas posted thisEveryone's saying cold email is dead and the fix is "more personalization" I disagree with both. Cold email with the wrong "targeting" is dead. No amount of personalization saves a message sent to someone who has zero intent. But cold email to someone actively experiencing the problem you solve? That still converts beautifully. The problem is your list! Stop obsessing over your subject line. Start obsessing over whether the person you're emailing actually needs what you have - right now, this week! That's the game. Has cold email worked for you recently? What changed when it did? Share below. #ColdEmail #SalesTips #HotTake #LeadGeneration #Jesse #OutboundSales Agent Jesse
-
Sudipta Biswas shared thisEveryone in prospecting is racing to add more filters. More dropdowns. More checkboxes. More ways to slice a database that was already stale the day it was built. We think filters are the problem, not the feature people are missing. A filter can only ask about things that don't change day to day - industry, headcount, title. It can't ask "who's publicly complaining about the exact thing we fix." Those aren't filters. They're moments. And moments live on the internet, happening right now. That's the actual difference between us and Apollo or Clay. Apollo gives you 8-10 filters to stack. Clay makes you wire together five data sources and hope they agree. Both are still fundamentally asking a static table a question, then handing you whatever was true when someone last refreshed it. Jesse doesn't have a database to filter. It reads the live web - permit filings, hiring posts, funding news, conference pages every time you search, so what you get back is what's true today, not what was true when a vendor last scraped it. We're not trying to build a better filter. We're trying to make filters unnecessary. Give it a try - https://agentjesse.ai/ Agent Jesse
-
Sudipta Biswas shared thisIt's been 30 days since Agent Jesse launched publicly. We opened it up to a handful of customers & design partners a few weeks before the launch. I wasn't sure what to expect. 30 days later, we are looking at 800+ signups 35+ paying customers Users across 50 countries The numbers are exciting. But the fun part is seeing how differently people use it. This made me think that we’re building something much bigger than another sales tool in the market. Still early. Still shipping every day. Still reading every piece of customer feedback ourselves and learning how we could make our product better. I'd love to hear what you think of Jesse or, if you've tried it already, what's one thing you'd like us to improve? https://agentjesse.ai/ Agent Jesse
-
Sudipta Biswas shared thisFloworks is hiring. And I'll be honest about what I'm actually looking for. Not someone who says they're a "content strategist" because they post videos on Youtube. I want someone who's already grown a YouTube channel. > It could be your own. > A client's. > A faceless channel. > A niche channel with 10,000 subscribers. > Or one with a million. If you've figured out how to get people to click, watch, and come back, I want to talk. We're hiring: YouTube Growth Consultant We're looking for someone who understands YouTube beyond editing videos. Someone who can: - Develop content strategies that actually grow channels. - Analyze retention, CTR, impressions, and audience behavior. - Improve titles, thumbnails, and packaging. - Identify winning content opportunities using data. - Turn good videos into channels people binge. What this isn't: - This isn't a role where you'll spend weeks making PowerPoints or sitting through endless meetings. - You won't be "supporting" someone else's work. - You'll own growth. What this is: A chance to work with Floworks, where we're building AI-powered products to automate sales. You'll work directly with founders, move fast, experiment constantly, and see the impact of your ideas in real numbers. If you've been quietly growing channels while everyone else was talking about growth... This is me noticing. To apply, comment or DM me with: - The YouTube channel you've grown (your own or someone else's). - What you actually improved (CTR, views, subscribers, watch time, retention, etc.). - The biggest lesson you learned from growing it. - No long resumes. - Just show me what you've built.
-
Sudipta Biswas shared thisBefore my team sends a single message, we ask three questions: ❓What just happened in this person's world? New funding? New hire? New product launch? Timing is everything. ❓Why would they care "today"? Generic pain points don't convert. Real-time context does. ❓What would make them stop scrolling? If your opening line wouldn't make YOU pause, rewrite it. This framework sounds simple. But most outbound fails at question one. We built Jesse(https://agentjesse.ai/) around the idea that the best time to reach someone is when they're already thinking about the problem you solve. What's your pre-outreach checklist? I'd love to compare notes. Agent Jesse #OutboundSales #SalesFramework #LeadGeneration #B2B #Jesse #FounderPlaybook
-
Sudipta Biswas shared thisWe launched Jesse on TinyLaunch this week. And we're currently sitting at #2 for the day 🚀 Right before this, we launched on Product Hunt. Hundreds of people signed up, and many of them are already using Jesse to find their next customers. The signups were exciting. But what really stood out was something else. People were opening Jesse, finding companies showing real buying signals, and reaching out within minutes. That was the bet when we started building Jesse. Not another database of millions of contacts. Not another tool that gives you a giant list of companies matching your ICP. We wanted to solve a much simpler problem: "Who has a reason to hear from us today?" Because outbound works a lot better when you reach out at the right moment. Now we're excited to bring Jesse to the TinyLaunch community as well. If you have a minute, we'd genuinely appreciate your support. And if you try Jesse, we'd love your honest feedback and your toughest questions. https://lnkd.in/gUzm7CFa Christopher Woggon
-
Sudipta Biswas posted thisThis is the single biggest GTM mistake Not building lookalike list. All of us must be having some customers. Some customers came inbound, few via referrals, few via word of mouth- it does not matter. What matters is that these customers are able to solve their pains with your product. Hence, once they are your customers- you should be "finding more of these". This is by far the most underrated GTM hack I have come across. There are few lookalike tools you can try- Ocean, Jesse (https://agentjesse.ai). While using these tools, you should be able to find out why certain new customers were surfaced- what were the logic behind these new suggestions, and then run a focussed outbound/ABM campaign on those list.
-
Sudipta Biswas liked this"We're paying this much for one meeting?" It's usually a sales or revenue leader asking. And I get why from where they sit, a meeting is one outcome. Someone replied, it's on the calendar, done. But they only see the last step. What they don't see: finding the right accounts, researching buyers, testing channels, writing outreach, following up, filtering replies, cutting poor-fit prospects, chasing no-shows. Most of that is invisible to the person reading the final number. So the meeting gets judged on its price tag, not on what it took to produce a good one. And the part that gets missed most: a cheap meeting with the wrong buyer isn't cheap. It costs the sales team an hour, kills momentum, and clogs the pipeline. That cost just never shows up on the invoice. So I don't ask, "How much did this meeting cost?" I ask, "Was it worth having?" Filling calendars is easy to measure. Filling them with the right conversations is what actually moves revenue.
-
Sudipta Biswas liked thisToday, my co-founder Manoj Krishnan and I are excited to introduce MachGen (https://www.machgen.ai/), an inference platform for diffusion and world models, delivering up to 6x faster generation while consistently operating in the lowest-cost tier. While LLMs have become dramatically faster and more efficient, diffusion models are still where LLMs were a few years ago: slow, expensive, and difficult to scale. Image and video AI are growing fast. But the infrastructure behind them hasn't kept pace. That's why we built MachGen. Our first independent benchmark results from Artificial Analysis found MachGen to deliver the fastest generation times across all four models they benchmarked, with image generation in as little as one second and video up to 6x faster than the next-fastest provider. Across every benchmark, MachGen also landed in the most cost-efficient tier — the only provider consistently in the fastest, lowest-cost quadrant. Not fastest on one workload. Fastest across all four. Better infrastructure isn't just about making models faster. It's about getting more performance from the GPUs developers already have. We believe that's what will make image and video AI practical to build and deploy at scale. We're encouraged by these early results because they reinforce what we've believed from the beginning: better infrastructure can make image and video AI faster, more efficient, and ultimately more practical to build at scale. • HiDream - https://lnkd.in/g5PXHTcm • Flux - https://lnkd.in/gWvYmpFe • Wan 2.2 - https://lnkd.in/ge2fm6uT • LTX - https://lnkd.in/g2PKSCM2 To learn more, visit https://www.machgen.ai Limited-time free credits and discounts!
-
Sudipta Biswas liked thisSudipta Biswas liked thisThis will be the game changer. We are Live on Product Hunt. Try Jesse: agentjesse.floworks.ai Check out the launch: producthunt.com Would really appreciate your valuable feedback.
-
Sudipta Biswas liked thisSudipta Biswas liked thisWe realised something uncomfortable. We weren't just competing with other prospecting tools. We were competing with a browser tab. Sales teams were already using ChatGPT to research prospects, understand accounts, and write emails. So we brought Jesse to where they already work. Today, Jesse is live on ChatGPT. You can now discover accounts, research prospects, and build outreach, all in one place. Fewer tabs. Less context switching. More selling. Want to try it? Drop a comment, and we'll help you get started.
Experience
-
-
-
-
-
-
-
-
-
-
-
-
-
Education
-
-
-
-
Publications
-
Traffic-Aware Green Cognitive Radio
Elsevier
See publicationCognitive Radio Network (CRN) has emerged as an effective solution to the spectrum under-utilization problem, by providing secondary users (SUs) an opportunistic access to the unoccupied frequency bands of primary users (PUs). Most of the current research on CRN are based on the assumption that the SU always has a large amount of data to transmit. This leads to the objective of SU throughput maximization with a constraint on the allowable interference to the PU. However, in many of the…
Cognitive Radio Network (CRN) has emerged as an effective solution to the spectrum under-utilization problem, by providing secondary users (SUs) an opportunistic access to the unoccupied frequency bands of primary users (PUs). Most of the current research on CRN are based on the assumption that the SU always has a large amount of data to transmit. This leads to the objective of SU throughput maximization with a constraint on the allowable interference to the PU. However, in many of the practical scenarios, the data arrival process of the SU closely follows an ON–OFF traffic model, and thus the usual throughput optimization framework may no longer be suitable. In this paper, we propose an intelligent data scheduling strategy which minimizes the average transmission power of the SU while maintaining the transmission delay to be sufficiently small. The data scheduling problem has been formulated as a finite horizon Markov Decision Process (MDP) with an appropriate cost function. Dynamic programming approach has been adopted to arrive at an optimal solution. Our findings show that the average transmitted power for our proposed approach can be as small as 36.5% of the power required for usual throughput maximization technique with insignificant increase in average delay.
Patents
-
SYSTEM AND METHOD FOR PROVIDING A BINARY OUTPUT BASED ON INPUT PACKETS OF DATA
Issued IN 344109
Courses
-
Communication Signal Processing Algorithm
-
-
Convex Optimization
-
-
Data Structure and Object Representation
-
-
Information Theory
-
Projects
-
Game Theory in Wireless Networking
- Present
Successfully implemented concepts from Game theory to derive an iterative waterfilling algorithm for noncooperative scheduling in heterogeneous networks. A new sequential optimization approach has been developed to solve an otherwise intractable game. The algorithm developed is a generalization of the famous Waterfilling algorithm that is applicable for a single player scenario. The convergence proof is under consideration.
-
Long Term Average Rate Optimization in Cloud Radio Access Network
-
This project centers around an upcoming communication environment, known as Cloud Radio Access Network (CRAN). In such a network, multiple Base Stations (BSs) cooperate among themselves in order to improve the coverage of the network. Optimal transmission strategy for one such downlink CRAN has been obtained in this project.
-
Interference Cancellation in Wireless Duplexer
-
An online signal processing block was designed for improving isolation in a duplex radio. The DSP block would employ the non-correlatedness of transmitted noise to estimate an unknown channel gain, which would otherwise require a pilot signal.
Honors & Awards
-
Institute Silver Medal
Indian Institute of Technology Kharagpur
Recipient of the Institute Silver Medal for being the topper of the class of 2016 in the Dept of Electronics and Electrical Communication Engineering
Languages
-
English
Full professional proficiency
-
Bengali
Native or bilingual proficiency
-
Hindi
Limited working proficiency
Other similar profiles
Explore more posts
-
Aditya Arora
Forget the Silicon Valley Playbook. The 'Blinkit Effect' is the real secret behind India’s next $50B+ operational explosion. While some look at Blinkit as just a delivery app, VCs look at it as an Operational SEAL Team. From the high-pressure dark stores of 2021 to the disruptive founders of 2025, if you want to see where India’s execution-heavy decacorns are coming from, you look at the Blinkit Effect Here are my 5 VC-level insights into why the "Gurgaon Mafia" is currently rewriting the Indian execution playbook: 👇🏻 1) Prithvi Singh - Gameskraft Most gaming founders focus on "fun." Prithvi used his Blinkit-honed high-concurrency roots to build a platform that handles millions of real-time transactions without a glitch. He leveraged the "Zero-Latency" mindset to solve backend engineering hurdles that slower competitors simply couldn't code their way out of ⚡ 2) Varun Khurana - Otipy Varun realised early on that Agritech isn't a "farming" business; it’s a density problem. By treating every tomato like a time-sensitive delivery packet, he solved the "perishability" trap in farm-to-fork. Otipy’s hyper-local supply chain is a direct product of the 10-minute delivery DNA native to the Blinkit war room 🥦 3) Zaheer Khan - Ayu Health Proving that you can bring "Quick Commerce" rigor to healthcare. Zaheer didn't just focus on "Doctors"; he focused on a systemic pain point: fragmented hospital operations. By bringing "Blinkit-level" logistics to hospital networks, he is systematising an industry known for chaos, demonstrating the "Systemic-First" ambition of the modern Blinkit graduate 🏥 4) Manas Gupta - GobbleCube Mastering "Data-Driven Velocity." Manas took the intimidating world of e-commerce brand analytics and made it as fast as a flash sale. By focusing on real-time SKU-level precision and scalable data frameworks, he successfully navigated the "Insight Gap" for D2C brands that were drowning in slow data 📊 5) Abhishek Upadhyay - CubeAPM The invisible backbone of the 2026 enterprise. By launching an observability platform for complex digital stacks, Abhishek is ensuring that "Zero Downtime" is a reality for the global market. He proved that the Blinkit network doesn't just build apps; they build the infrastructure that keeps the world’s most complex systems running 🧊 The ➡️ Insight: The "Blinkit Effect" isn't about the bike or the bag; it’s about Urgency as a Culture. When your training ground is a 10-minute delivery window, your baseline for "fast" shifts. That psychological edge is what creates a high-velocity, $50B+ ecosystem. ➡️ The 2026 Blueprint: Don't just look for a good idea; look for an Execution Moat. The next unicorn won't just have a great product; it will have a founder who can survive the high-pressure "Blinkit" crucible and scale at 10x speed 📈
7 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content