Aparna Sinha

Fundamental opportunities worth solving for in AI right now: Predictability of reasoning, planning, execution of AI. Efficiency and ease of use of running AI workloads. Multiple system integrations - from DBs and APIs to infrastructure and biz systems. Trust, security, guardrails, authorization and auth. Business/science domain specific data, workflows, and optimization. AI cost understanding, control, and reduction.

30

5 Comments

In the fast-paced world of AI, leaders like Sharath Rajasekar, CEO of Javelin (www.getjavelin.com), inspire us to push boundaries while prioritizing security. Under his visionary guidance, Javelin is revolutionizing how enterprises deploy AI agents safely and efficiently. Thrilled to see Javelin's latest announcement: MCP Security, a game-changing defense-in-depth solution for the Model Context Protocol. It features Ramparts for pre-deployment risk scanning against supply chain threats and poisoning, plus Runtime Guardrails for real-time policy enforcement—blocking vulnerabilities without hindering innovation. Javelin's commitment to reducing risks, accelerating AI rollouts, and ensuring compliance sets a new standard. Congratulations Sharath Rajasekar and the entire Javelin team! Read more: https://lnkd.in/ea2eM9hM

43

1 Comment

GPUs have gotten faster, and models have gotten larger. But storage is still stuck in the past. Modern AI workloads need instant access to huge datasets. Instead, teams waste time on slow retrieval, idle GPUs, and costly pipelines. Archil is changing this. It delivers object-storage scalability at block-storage speeds. The platform allows for seamless, high-performance access to massive datasets. By eliminating cold starts and accelerating throughput 30X, Archil enables teams to fully utilize their infrastructure and move faster on training, analytics, and AI deployment. Archil’s founder Hunter Leath brings rare technical depth and operating experience from building Amazon EFS and optimizing Netflix’s cloud performance. This is a team building the foundational layer that AI workloads desperately need. I’m thrilled to lead Archil’s seed at Felicis with Nancy Wang and support their mission to reinvent cloud storage for the AI era. https://lnkd.in/gWueXhzz

65

2 Comments

Today on Training Data: Sanjit Biswas, founder & CEO of Samsara (NYSE:IOT) and former Sequoia Capital-backed founder of Cisco Meraki Sanjit shares the ups & downs of running neural nets on constrained compute and power footprints in the real world, ~2-10 watts Physical AI is hard 🫡 Listen to the full episode: Apple: https://seq.vc/mze Spotify: https://seq.vc/zq3 YouTube: https://seq.vc/185b91

72

4 Comments

The next frontier of AI isn't just about choosing the right model. It's about building moats through implementation. Across the Emergence Capital portfolio, companies like Bland and Federato are scaling faster by embedding Forward-Deployed Engineers. These engineers bridge the gap between AI and reality, ensuring real results in production. Read more from my colleagues Gordon and Kyle on why FDEs are the key to unlocking AI's full value. https://lnkd.in/gcsRRGX2

29

3 Comments

On-Prem is the New Cloud: Why Enterprises are Reversing Stance in the Age of AI Remember when cloud was the battleground for innovation, agility, and scale? Everyone was tracking the % of on-prem compute moving to the cloud and it was growing every year? Well now there’s a reversal. The reality today? On-prem is the new cloud. Here’s why. AI’s Dark Side: Legal Risk and Data Governance Headaches Enterprises are excited about AI, but also uneasy as vendors like OpenAI face intensifying legal pressure. In a high-stakes lawsuit, a judge ordered OpenAI to preserve all ChatGPT chats (even deleted ones), overriding its 30-day deletion policy. Imagine that: you delete sensitive conversations, but they’re still sucked into legal limbo. The precedent is clear: AI prompts and outputs are discoverable records, demanding integration into enterprise ESI (Electronic Stored Information) policies. For large organizations handling sensitive data, the risk is simple: you can’t afford your AI tools to turn into legal liabilities. Especially when global regulations (GDPR, etc.) and internal governance collide with unpredictably broad discovery orders. Sovereignty, Control, Trust Many enterprises still ban ChatGPT or impose strict rules on usage. Even when allowed, token limits are throttled. Enterprises are cautiously letting AI in (as they should). Enterprises need to have control and trust. SAP is leaning into this with its Sovereign Cloud On-Site, deployable on customer premises with SAP-managed infrastructure. It is built to deliver data, operational, technical, and legal sovereignty in one package. Doubt it will be the last. Scaling Pains Talking to founders, three reasons stand out why cloud isn’t scaling with AI: 1/ The rate at which AWS/GCP lets you provision compute and the compute it wants you to provision both falter as AI and AI driven developers deploy a lot more code more frequently than before. So sometimes it’s not about data governance, but actually about performance and functionality. 2/ Shadow AI is rampant—employees paste sensitive info into models despite restrictions—creating demand for better permissioning tools. 3/ Data leakage remains a live risk, like the August 2025 Grok incident where hundreds of thousands of private chats became publicly accessible and indexed. One thing is for sure - founders selling into the enterprise are scaling faster if they offer the ability to run on-prem. The Future is Hybrid Don’t get me wrong - cloud is not going anywhere. But neither is on prem. A few years agoI thought we were headed for a full cloud shift, but now I think the future is truly hybrid.

34

4 Comments

The next great AI race has begun, and it’s not about compute or talent. It’s about data. We spent $400 billion this year on GPUs and billions more competing for AI talent. But the investments that matter most now are in data quality. The difference between economically useful AI and “slop” isn’t teraflops or PhDs. It’s expertise. Early AI models scraped the open internet, learning from keyword-stuffed posts and half-truths. They became impressive generalists but poor economists, consultants, or engineers. Real economic value comes when models learn from experts: the people who actually know how to make things work. That’s where Mercor comes in. Mercor works directly with experts to learn how they think and make decisions, and then trains AI on that knowledge. It’s transforming lived expertise into structured learning for machines. The shift is already massive. The enterprise SaaS market is worth $1.5 trillion, but even that’s just a fraction of corporate operating spend. AI is poised to double or triple that number by making labor (roughly 80% of business costs) more valuable and scalable. Mercor is rapidly becoming the place where AI learns to do real, economically valuable work. The first ten-trillion-dollar company may well be built not on silicon, but on expertise. Mercor crossed a $500M revenue run rate this September and is profitable, with zero enterprise churn. We at Felicis are thrilled to invest in their Series C because they are building the next generational platform technology company. Mercor is re-defining the horizon of expertise: unlocking not only corporate value, but potentially the next chapter of human progress. Read more from Mercor: https://lnkd.in/gPpqCrBq More from me on X: https://lnkd.in/gc-2w_6h

256

8 Comments

The most promising founders in stealth in 2025. We used Crustdata to search for stealth founders with traditionally impressive backgrounds and here are the results: Sandeep Kachru: Previously head of Data platform at Airtable and data engineer at Facebook and Google. Rohan Shah: Founder of 3 startups and previously at McKinsey. University: University of Pennsylvania Lawrence Ibarria: Founder of 2 startups in the robotics space and previously was an Associate Director at McKinsey. University: Georgia Institute of Technology Arthur Dubois: Previously was VP of engineering at Xwing (acquired) University: Stanford and McGill Theophile Gervet Previously was a Research Scientist in Mistral AI and held lead/senior positions in Skild AI and Relyance AI. Adriane McFetridge: Previously was a Director of Engineering at Netflix. University: University of Texas. Vallari Mehta: Previously was a ML engineer at Lyft, Criteo and Veritas tech. University: Carnegie Mellon. Obafemi A.: Previously worked as an Investor in Lead Edge Capital and has a wide range of internship experience from Uber to General Atlantic. University: Stanford and Texas McCombs School of Business. Connor Owens: Has a wide range of experience at big name companies such as Google, YouTube and Meta. University: Stanford School of Business and Northwestern University. Michael Pao: Was previously the head of product at Uber and the cofounder of Trove (acquired). University: Harvard Business School and University of Pennsylvania

153

10 Comments

In the latest episode of the OpenAI podcast I got talk to OpenAI CFO Sarah Friar and legend himself Vinod Khosla about the AI industry. Sarah explained the relation between compute and revenue (OpenAI's revenue keeps 3x'ing each year) and Vinod provided an insightful way to tell if we're in a bubble: API calls. Are they going up or down? (Spoiler: Up, up, up!) https://lnkd.in/g2tTQBdp

22

1 Comment

Garry Tan recently mentioned that forward deployed engineers at Palantir are the best candidates for starting vertical AI companies, and I completely agree. The deep customer engagement and knowledge about opaque enterprise workflows is crucial in this case. And guess who has been forward deploying engineers in US enterprises for the last 40 years? The IT services industry in India. This deep knowledge about enterprise workflows combined with an extensive bench of entrepreneurial talent in India makes India a fantastic source of vertical AI companies. We are already seeing this happen and will see more and more of this in the coming decade.

156

7 Comments

The race for the AI future isn’t really about more computing power or better large language models, as important as those things are.    The true revolution will happen in AI’s interactions with what I call nature’s intelligence – the logic of the natural world, evolved over billions of years – and human intelligence.    Nowhere is better positioned to facilitate those interactions and lead that revolution than Massachusetts, with its world-leading life sciences ecosystem, highly educated workforce, and burgeoning AI sector.    I shared some thoughts about how my home state can lean into the uncertainty it’s facing and win the AI future in Boston Business Journal. https://lnkd.in/e7AtE93F

365

8 Comments

Most of the AI tooling conversation right now centers on code generation. Cognition, Cursor, Claude Code....these tools are genuinely useful, but engineers are drowning in production issues. At our latest Greylock Partners Change Agents event, Mayank Agarwal and Rushin Shah from Resolve AI walked through a reality most engineering teams live through: with AI coding tools "you can crank out code much faster, but teams aren't really shipping to production that much faster." Some takeaways from our conversation: 1/ The real problem isn't code generation. Mayank saw this at Splunk where engineers spent only 10-20% of their time writing new features. The rest was fighting incidents, navigating tribal knowledge across hundreds of microservices, and operating tools that don't talk to each other. 2/ Production systems are different from code. Your codebase is static. Production is emergent behavior from infrastructure, deployments, traffic patterns, to tribal knowledge buried in Slack. As Mayank stated, the issue is "a coding agent cannot predict these things without the deep understanding of what's actually happening in production." 3/ The scale required is massive. Resolve is now consuming a trillion tokens per quarter—1% of all Azure usage. That's running Stack Overflow from scratch twice per week. "Vibe debugging" is the missing piece, Rushin's term for getting clarity on production systems the way vibe coding accelerates writing code. Check out the full episode in the comments

104

28 Comments