About
In his role as the global lead for Google Cloud’s Partner GTM Practices, Pallab is…
Activity
10K followers
-
Pallab Deb posted this#FDEs worked for Palantir Technologies because it was all about moving data to #Foundry and surfacing insights from that data - the expertise and engineering centered on #Foundry - a wholly owned Palantir Technologies product.. #AI #Agentic workflows are about working and interfacing with legacy systems and complicated business processes... this calls for deep knowledge of the business process and deep awareness of legacy systems of record and engagement complemented by full-stack #agent #engineering skills ... how will #FDEs born into a frontier model ecosystem develop breadth of expertise and depth of skills to delivery the promise of #agentic #ai on top of a legacy reality? Thought leaders - WDYT?
-
Pallab Deb reposted thisAt CapitalG, we’ve been fortunate to back generational companies such as Stripe, CrowdStrike, Duolingo, and Databricks, and to support them with both capital and value-added expertise. Today, we are thrilled to add Isomorphic Labs to this remarkable cohort of companies—and Sir Demis Hassabis and Max Jaderberg to the list of visionary leaders—with whom we’re fortunate to partner. Laela Sturdy,Jill (Greenberg) Chase, Manmeet Gujral, Nicholas Kim, and all of us at CapitalG are incredibly proud to back Isomorphic Labs in this funding round as their world-class team and category-defining AI drug design engine transform the future of medicine.Pallab Deb reposted thisToday marks a pivotal moment for Isomorphic Labs. We have secured $2.1 Billion in our second external funding round, led by Thrive Capital with significant participation from Alphabet and GV. They are joined at the table by new investors MGX, Temasek, CapitalG and the UK Sovereign AI Fund. We're thrilled to have their combined expertise on board. This milestone funding accelerates our ability to develop the pioneering novel AI models that power our AI drug design engine (IsoDDE) and deploy them at scale: accelerating and expanding our pipeline of therapeutic programs driving toward the clinic. By transforming drug discovery with proprietary AI at the core, we are positioned to deliver scientific breakthroughs with a level of precision that was previously unthinkable. To power this next phase of growth, the funding will also support scaling our expertise; integrating world-class AI, engineering, drug design, and clinical expertise across our global sites. This scale is crucial to our long-term vision of applying AI-driven breakthroughs to the most complex biological challenges and addressing the global burden of disease. Our goal is to bring life-changing medicines to patients who need them most. We are grateful to our incredible team, partners, and investors who share our ultimate mission: to one day solve all disease with the help of AI. Read the full announcement here: bit.ly/4v2OI03
-
Pallab Deb shared thisAlways exhilarating to jam with R "Ray" Wang on the state of the #AI #Union. To win in this 2026 landscape, you need more than just tools—you need the #Winning #Trifecta: 🏭 Domain Expertise (The Context) 🧠 Precision Data, Agent Harnesses and Models (The Intelligence) 🔋 Compute & Power (The Engine) But the real differentiator? A #Silicon #Valley mindset that prioritizes #10X velocity over incrementalism. Great to dive deep with the best in the business. 🚀
-
Pallab Deb reposted thisPallab Deb reposted thisJeff Dean and I had the pleasure of hosting Dina Bass at our Mountain View campus recently for an in-depth conversation on Google’s approach to AI infrastructure. We shared the history that led us to engineer entire full-stack systems, instead of simply buying off-the-shelf components, including the insight a decade ago that led us to create our own custom silicon: that we wouldn’t be able to provide language translation and voice recognition otherwise. It turned out that there was a lot more that custom silicon would enable us to do as Transformers came along a few years later. AI workloads require massive mathematical calculations, and these early revelations around what it would take to train and deploy AI at scale helped set the standard for AI-optimized hardware. Our teams continue to partner deeply to co-design hardware to essentially predict the future needs of researchers lined up with hardware design and delivery cycles. A few key themes from our discussion: The Co-Design Advantage: Our AI & Infrastructure (AI2) and DeepMind teams work as direct partners in the same room. This closed-loop system allows us to both set the design and to use AI to design the physical layouts of the very chips that will train the next generation of models. Systems, Not Just Chips: There’s a lot more that goes into powering intense AI workloads than chips. By using proprietary networking fabric and optical circuit switching (OCS), we enable thousands of TPUs to act as a single supercomputing "brain." Bending the CapEx Curve: By controlling the design from silicon to the facility, we eliminate lost efficiencies at the interfaces between hundreds of hardware and software components. This allows us to reinvest savings into large-scale research and lower the marginal cost of training and serving models for our customers. The research-to-hardware flywheel is an accelerating feedback loop. I am incredibly proud of how our teams are redefining what is possible by building infrastructure that is fundamentally aware of the models it supports. Stay tuned for more exciting news to come at this year’s Google Cloud Next! #AIInfrastructure #GoogleCloudNext #CloudInnovation https://lnkd.in/g3NwDTzmGoogle Eyes New Chips to Speed Up AI Results, Challenging NvidiaGoogle Eyes New Chips to Speed Up AI Results, Challenging Nvidia
-
Pallab Deb shared this#Experience #Certainty and #Innovation with Tata Consultancy Services and Google Cloud as you get your #enterprise ready for #AI adoption at scale. Fragmented data is almost always the biggest roadblock on this journey. We will be addressing how both firms jointly address this area with an #AI first approach to creating the #unified data platform to drive your #AI flywheel Siddhartha Gupta Ranjith Jayaraman Aneesha Chhikara Yasmeen AhmadPallab Deb shared thisJoin us live at #GoogleCloudNext ’26 as leaders from Google Cloud and Tata Consultancy Services discuss how enterprises are accelerating data modernization using #agenticAI to transform legacy data estates into scalable, AI-ready foundations that drive faster time-to-value and smarter decision-making. Featuring: Siddhartha Gupta | Pallab Deb Tata Consultancy Services - North America | Tata Consultancy Services - UK and Ireland | Tata Consultancy Services - Europe | Google Cloud Partners #TCSandGoogleCloud #TCSatGoogleCloudNext
-
Pallab Deb shared thisHari Bhaskaran - love your grounded commentary and subtle humor in your updates from the many experiments you run on Claws, Models, Frameworks and AI Infra. Keep them coming...Pallab Deb shared thisCompared Ollama+MLX+Qwen (released 4 days ago), Gemma4 (31B) released today, against Claude Opus. The first two are running local. TL;DR - Don't cancel your Claude subscription yet, but getting close!. For OpenClaw usage, both Gemma4 and the Qwen one look good enough. Doing finance / stuff you don't want to send to cloud would be perfect use case for these ones. - Test 1 (simple): Produce a report from a CSV file. Something like asking your OpenClaw to look at credit card statements or Strava results. - Test 2 (complex): Dependency resolver for a task graph — topological sort, parallel execution, cycle detection, critical path. Qwen 35B (thinking off): With thinking on, Qwen spent 10 minutes on Task 2 and produced zero code. With `"think": false`, both tasks completed in seconds. Task 1 was a tie. Task 2 had a bug. But this might still be good enough for typical personal-agent math - reports, summary etc. Gemma4 31B (thinking on): Completed both tasks & no timeout issues (unlike Qwen). Task 1 gave similar code. Task 2 was bug-free with the cleanest code of all three models. Tradeoff: 72s vs Opus's 17s
-
Pallab Deb shared thisYou nailed it Nithin Kaimal - 5% is likely the ceiling - unfortunately for them, not the floor.. painful restructuring can raise the ceiling but the fuse is short. Critical for customers, innovators, investors and product principals alike to factor in as to where they place their bets. Bharat Ram Ivaturi Bhanumurthy Ballapuram Kevin Ichhpurani Rob BirdsongPallab Deb shared thisMehul Gupta and I analysed the ten NIFTY IT companies + Accenture Cognizant data over the last 5 years (in USD terms, to remove USD appreciation noise) to understand what's happening with their productivity. Two numbers stood out: - Revenue per employee: ~3% CAGR - EBITDA per employee: ~1.6% CAGR More recently, EBITDA per employee is up 5% year-over-year for last year - a meaningful acceleration, driven largely by headcount optimisation and operational efficiencies. But the “5%” number is really important if you're a founder building AI-first services: That 5% improvement represents the ceiling of what large incumbents can currently achieve through incremental change. They're operating with lakhs of employees, legacy business models, and deeply entrenched incentive structures. Transforming that DNA is extraordinarily hard. This is exactly your opportunity as an AI-first services company today: - You can target disproportionate revenue growth per employee from day one - You can build towards software-like gross margins (75%+) instead of the industry's ~40% - You can continue to grow exponentially, with minimal external funding The CEO of a leading IT services company aptly highlighted to me that "True AI-first companies will deliver disproportionate growth in revenue and EBITDA relative to headcount, with software-like margins." The gap between incumbents improving incrementally and AI-native companies building from scratch isn't a criticism of traditional IT. It's a recognition that large-scale organisational transformation is extremely difficult and how that difficulty creates space for new category-defining companies.
-
Pallab Deb shared thisSuch a rich and fun conversation Alyssa Cox and Nidhi Srivastava ... I have never had so much fun being on a panel! :) Alyssa Cox - looking forward to having you share the story and more at #GoogleCloudNEXT Sanjeev Vohra Riju Vashisht Kevin Ichhpurani Chris Sakalosky Michael Clark Aimee CatalanoPallab Deb shared thisThrilled to have hosted the fireside chat to spotlight on #CardinalHealth AI Transformation journey with #Genpact and #GoogleCloud at our flagship event #FOCUS26 in Kolkata. What a vibe – diving into #digital transformation, #AI strategies for scale, and the dynamics of #ecosystem partnerships with Alyssa Cox, VP of Transformation at Cardinal Health and Pallab Deb, MD of SI and Industry Partnerships at Google Cloud. We explored three things – a) Design patterns for AI scaling b) Importance of process intelligence and a solid data foundation to build the transformation roadmap c) The power of the ecosystem partners and how to bring the best to the client. Key takeaways from our session: 1.#Be #intentional about having an #AI scaling strategy 2.There is a lot of noise and hype; focus on #process and #data to build sustainable value 3.The partner ecosystem is the new village for #transformation and success A big shoutout to Alyssa and Pallab for joining me as panelists. The candor, wit and humor made for a sparkling discussion! And the BTS conversations and laughter was the bonus 😊. #Genpact #GoogleCloud #FOCUS26 #AgenticAI #EnterpriseTransformation Alyssa Cox Pallab Deb Kevin Ichhpurani Victor Morales Samrah Khan Sanjeev Vohra Riju Vashisht Anil Nanduru Jinsook Han Caitlin Blewett Yugal Joshi Ojas Save Shalvi Chitkara Saipriya Natarajan Rajesh Ganshani Chad Stiller Manu Bhog Saurabh S. Dipty Ranjan Padhy Kiran Chadalavada Praveen Chilakalapudi Hemant Bhardwaj Kannupriya Meyyappan Tom R. Nuno Lopes
-
Pallab Deb shared thisSuch a rich and fun conversation Alyssa Cox and Nidhi Srivastava ... loved sharing the story of how Cardinal Health is driving #AgenticAI driven #Finance #Transformation with Google Cloud and Genpact, with 250+ business leaders in the historic city of #Kolkata... Sanjeev Vohra Riju Vashisht Balkrishan "BK" Kalra Matt Renner Kevin Ichhpurani Chris Sakalosky Oliver Parker Michael Clark Aimee CatalanoPallab Deb shared thisThrilled to have hosted the fireside chat to spotlight on #CardinalHealth AI Transformation journey with #Genpact and #GoogleCloud at our flagship event #FOCUS26 in Kolkata. What a vibe – diving into #digital transformation, #AI strategies for scale, and the dynamics of #ecosystem partnerships with Alyssa Cox, VP of Transformation at Cardinal Health and Pallab Deb, MD of SI and Industry Partnerships at Google Cloud. We explored three things – a) Design patterns for AI scaling b) Importance of process intelligence and a solid data foundation to build the transformation roadmap c) The power of the ecosystem partners and how to bring the best to the client. Key takeaways from our session: 1.#Be #intentional about having an #AI scaling strategy 2.There is a lot of noise and hype; focus on #process and #data to build sustainable value 3.The partner ecosystem is the new village for #transformation and success A big shoutout to Alyssa and Pallab for joining me as panelists. The candor, wit and humor made for a sparkling discussion! And the BTS conversations and laughter was the bonus 😊. #Genpact #GoogleCloud #FOCUS26 #AgenticAI #EnterpriseTransformation Alyssa Cox Pallab Deb Kevin Ichhpurani Victor Morales Samrah Khan Sanjeev Vohra Riju Vashisht Anil Nanduru Jinsook Han Caitlin Blewett Yugal Joshi Ojas Save Shalvi Chitkara Saipriya Natarajan Rajesh Ganshani Chad Stiller Manu Bhog Saurabh S. Dipty Ranjan Padhy Kiran Chadalavada Praveen Chilakalapudi Hemant Bhardwaj Kannupriya Meyyappan Tom R. Nuno Lopes
Pallab Deb commented on a post
20h
Our industry is known for jargon... On the ground reality of deployments makes the term irrelevant for the reasons you mention (deterministic, workflows etc)... So yes #AI native is another jargon but I consider that to allude to companies that are built on foundation models e.g. Cursor just like Cloud/Digital native were companies built on Cloud and didn't exist before that e.g. Salesforce
Pallab Deb commented on a post
1w
Congrats Manish Govil - super happy to see you having fun and making an impact.
-
Pallab Deb liked thisPallab Deb liked thisFirst - thanks again for the 1000 or so folks who connected about the job posting I did. I only need a handful of people but I am quite grateful for your response. Our recruitment team is working on next steps. In the process of going through the responses - I felt it might be a good idea to give a high level idea of what I look for when I hire these roles. It's not an exhaustive list by any stretch but hopefully it gives your a rough idea of whether it suits your skills and aspirations. https://lnkd.in/d6wM6tWVWhat I look for when hiring engineers, engineering managers, architects and product managersWhat I look for when hiring engineers, engineering managers, architects and product managers
-
Pallab Deb liked thisPallab Deb liked thisMy thoughts on rest of 2026 :) 1. 𝐀𝐩𝐩𝐥𝐞 will resurrect or complete their rumored Project ACDC to build a server product that competes directly with the Intel/AMD/NVIDIA server-monopoly. Power advantage + Unified memory alone will justify switching over to Apple. The only things stopping them are memory shortages, CUDA, and them not wanting to share. I think they will hold it for internal use only to take a cut of the hyper-scalers' profits. In that world, 'Apple Infra' will give low-power-high-memory AI-on-rent while others can't. If they did sell, they would have to open up macOS to the cloud providers more - why bother? 2. 𝐌𝐢𝐜𝐫𝐨𝐬𝐨𝐟𝐭'𝐬 Windows division has maybe 1 year of time left to rebuild an OS with more stability and a less insulting CLI world. Abandon WSL and probably clone BSD like Apple did many years ago. I say "It is never too late to clone a BSD". Have your AI agent build the user-land right. This is for server/developer only. Consumer market is already unrecoverable. NVIDIA-compatibility + CUDA + secure-os-for-agentic-actions play can help them get a hold in server market. People use Linux, not because it is great, but because they can't wait 10 years for the BSD world to get it right. If Microsoft is not planning this, they might as well close shop on Windows now. The Apple Neo from the consumer side and TUI-agents from the Developer side are screwing them simultaneously. HP/Dell/Lenovo/Asus etc might abandon/close their consumer business at any time. They will continue to make money from the cloud business, but once Apple's AI-on-rent materializes, it will kill the room for the 3rd cloud provider. AWS has market share, Google has their own AI-chip to survive, Oracle etc. have contracts and "contacts". Would be great if they stopped JS-Washing their apps. Office is on thin ice with Agents. Maybe it will manage to survive.
-
Pallab Deb liked thisSearce Inc is getting started to automate, improve & transform business processes with agentic AI at a scale never imagined before. Searce #fds (forward deployed solvers) enabling process-first transformation at the world's largest hybrid solar and wind plant - five times larger than the city of Paris and larger than the entire country of Singapore. #futurifywithsearce Hiring for strategy consultants, product managers & associate product managers with a deep tech engineering background for locations: Ahmedabad, Mumbai, Singapore and more. Ankit Sinha Vishal Jarsania careers@searce.comPallab Deb liked thisForward Deployed Solvers, quite literally. For years, Searce Inc was delivering Business Outcomes - and now our Engineers and Consultants are 10x more competent, 10x more enthusiastic. Clients appreciate when we brave the harshest of environments together with their teams. And who knows, we would also have done our bit to bring in Kardashev 2 by focusing on the right problems. It works on my system --> It works on the field --> It works on the balance sheet --> It works for humanity!
-
Pallab Deb liked thisPallab Deb liked thisWhat's with this narrative of "the world has moved from mobile native to cloud native to now AI native?" We certainly didn't throw away mobile or cloud when we started using AI. If anything we just added one more useful layer to architecture to make use of a novel technology I also find it frustrating when people say we have moved from deterministic systems to probablistic systems. We haven't done anything like that - we just changed how the logic gets executed to make the best use of deterministic and probablistic solutions. A good example is the classic screen sequences in a workflow. AI can make great decisions to use context and show you the most optimal flow - but this does not eliminate the need for rigid workflows . There are cases like compliance workflows where variability is not a good thing. Also for people who do repetitive work - a predictable sequence is better for efficiency. Or take any enterprise workflow in finance. You cannot risk hallucinations in the least - so you need to put a lot of deterministic guard rails around the probabilistic AI tools to get results you can live with Rant over :)
-
Pallab Deb liked thisPallab Deb liked thisQuantiphi turns Thir-TEEN! 🎉 13 years of Solving What Matters! From bold ideas to real-world impact, this journey has been shaped by the many ways we approach challenges—through diverse perspectives, out-of-the-box thinking, and that unmistakable Quantiphi spirit. Here’s to the passion, innovation, and togetherness that make Quantiphi what it is today!✨ #Quantiphi #FoundationDay #HappyBirthdayQuantiphi #SolvingWhatMatters #LifeAtQuantiphi
-
Pallab Deb liked thisPallab Deb liked thisThis week, I close an important chapter with Deloitte and prepare to begin a new one as Chief Practice Officer for the Financial Services Industry vertical at Fractal AI, which recently became a public company. My journey with AI started long before AI became the topic in every boardroom. In 1992, while studying mechanical engineering, I received a scholarship from the mathematics department (!) to research AI. More than three decades later, I am still at it, but so much has changed! As I transition into Fractal, I am reminded that clients are not just trying to understand AI as a technology shift. They are trying to understand what it means for their teams, institutions, customers, their children entering school or the workforce, the future of their industries and even mankind. The work ahead is not simply about deploying Fractal leading AI solutions. It is about helping leaders build confidence, challenge assumptions, move past uncertainty and fear, and turn this revolutionary tech into practical, scalable, positive action. I am grateful for my Deloitte AI chapter, for the clients and colleagues who shaped it, and for the opportunity ahead. The next chapter begins Tuesday.
-
Pallab Deb liked thisPallab Deb liked thisSankar Venkatraman has been elevated to Senior Vice President and Global Head of Enterprise Applications at Hitachi Digital Services, marking another significant milestone in his distinguished leadership journey spanning more than three decades across global technology and digital transformation organizations. In his new role, he is expected to lead enterprise application strategy, drive innovation, and strengthen digital transformation capabilities for global clients. Prior to joining Hitachi Digital Services, Sankar served as Principal Architect at solutions by stc, where he worked closely with enterprise CXOs on AI-driven transformation, digital platforms, cloud modernization, and large-scale IT architecture initiatives across Saudi Arabia. He has also held senior leadership positions at DETASAD, Wipro, HCLTech, Mphasis, Satyam Computer Services Ltd, and PwC LLP, leading multi-million-dollar businesses across enterprise applications, cloud platforms, SAP, Oracle ecosystems, and IT transformation. Known for his expertise in digital innovation, enterprise modernization, and strategic leadership, Sankar brings deep global experience to his expanded role at Hitachi Digital Services. #DigitalTransformation #EnterpriseApplications #Leadership #cxolanes
Experience & Education
-
Google
View Pallab’s full experience
See their title, tenure and more.
or
Honors & Awards
-
Best Practice Award
Annual CEO Awards, Wipro Technologies
-
Best Service Line Sales Manager
Wipro Technologies
-
Bravo Award for winning the largest deal in North America
Wipro Technologies
-
Member of All Star Team
Wipro Technologies
-
Best New Practice Manager globally
Wipro Technologies
Recommendations received
3 people have recommended Pallab
Join now to viewOther similar profiles
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content