Podcasts

Conversations with interesting people in AI. For technical deep dives and more content, check out Ground Zero

A Quest to Formalize Intelligence | Ludwig (Principal Engineer & Independent Researcher)

This is Episode 15 on Ground Zero! Ludwig is 31yo programmer with no formal education. He dropped out in 7th grade - started programming around age 10 - began working at 15 - and later worked on distributed systems for a company with hundreds of millions of users. You might know Ludwig for systems programming, ML, intelligence, mathematics, software-engineering culture and his X presence. Listen on Spotify: https://open.spotify.com/episode/0ApItqzYZfB4mpIeTt204B?si=gFLrMsCgTV6lGTpZ9gvMkQ Listen on Apple Podcasts: https://podcasts.apple.com/in/podcast/groundzero-ai-talks/id1896715519?i=1000769795467 ----------- Ludwig [Guest]: https://x.com/ludwigABAP Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://groundzeroai.in ------------ SPONSOR: Haize Labs (https://haizelabs.com) is building expert-level agents for mission-critical work, powered by proprietary Reliability Harness. They are doing some ambitious and interesting work. Checkout Leonard Tang (Cofounder and CEO) articles to learn more: https://x.com/leonardtang_/articles To sponsor future episodes, visit https://www.groundzeroai.in/partner ------------ TIMESTAMPS: 0:00:00 - INTRO 0:01:06 - What are you most excited about these days? 0:03:31 - All arcs are isomorphic, Dropping out in 7th grade 0:09:31 - Does skipping formal education make you naturally broader? 0:10:11 - Growing up on 4chan, Mathematics is humbling 0:14:52 - Mentors, First programming job, Internet microcultures 0:21:47 - Culture in SF, Why he won't start a company 0:25:21 - Does AI make self-teaching easier or trap you in infinite explanations? 0:29:49 - The 20% doing 80% inside a company 0:33:24 - Company culture and Building subcultures 0:35:56 - How do you define understanding, compression equals prediction 0:45:44 - Understanding vs knowing, Michael Levin, Cognitive light cones 0:54:39 - Starting a research lab, Sheaf theory and Grothendieck 0:58:43 - Mapping Levin's biology onto proof space with MCTS and Lean4 1:02:49 - Active inference, Distributed systems and Morphogenesis 1:12:28 - Signal vs noise on X, The unit distance proof, Flanderization 1:17:39 - Community Questions 1:21:39 - Shape rotator vs wordcel, Tenstorrent, How to hire great engineers 1:28:01 - Building mental space to do things that will be left to do 1:33:42 - Advice to a 20-year-old. Be a capable thinker, you have 20,000 days left

1.0k views • May 27, 2026

1.14 hours

The Delta of Intelligence is Human Data | Curtis Northcutt (Director of AI Research, Handshake)

Curtis Northcutt joins us on GroundZero. He is Director of AI Research at Handshake, founder of Cleanlab (acquired by Handshake) and invented confident learning during his PhD at MIT. In this conversation, we dig into why the delta between model generations is almost entirely human data, the four quadrants of the AI market, how expert data labeling is deeply misunderstood, the long tail of human knowledge that AI still can't touch, RL environments, coding benchmarks and nuances around Human Data market. PS: Handshake’s revenue has risen to nearly $1 billion up from $550 million in January and $5 million a year ago. Checkout: https://joinhandshake.com/ ----------- Curtis NorthCutt [Guest]: https://x.com/cgnorthcutt Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://groundzeroai.in ------------ SPONSOR: Haize Labs (https://haizelabs.com) is building expert-level agents for mission-critical work, powered by proprietary Reliability Harness. They are doing some ambitious and interesting work. Checkout Leonard Tang (Cofounder and CEO) articles to learn more: https://x.com/leonardtang_/articles To sponsor future episodes, visit https://www.groundzeroai.in/partner ------------ TIMESTAMPS: 0:00:00 - INTRO 0:02:08 - Snowman effect: Growing up in rural Kentucky to 7 years at MIT 0:09:50 - Confident Learning, Two Systems of Intelligence, Classical ML has to prevail 0:20:37 - Origin of CleanLab and Handshake acquisition 0:28:00 - How does GPT-5 become GPT-6? 0:31:20 - The Fastest growing Data Lab: What makes Handshake different? 0:35:32 - Difference between Good Data and Bad Data! 0:38:17 - New kind of Data, IA and AI 0:42:04 - Scaling Coding Benchmarks and Efficiency in Long Horizon Tasks 0:49:12 - On Pre-Training, Misconception around Human Data 0:57:13 - Open Question: Taste and Personality, Quick Fire 01:05:15 - Advice to 20yo : Skill and Obsession

495 views • May 9, 2026

1.56 hours

The Never Ending Lore of Harness | Vivek Trivedy (Product Lead, Langchain)

In this episode, Viv joins us on Ground Zero. He leads open source agents and harnesses at LangChain. In this conversation, we go deep into what a harness actually is, Deepagents, Context rot, Why agents should be more opinionated, Harness as a Service, File System, how different frontier models behave in agent workloads, Meta Harness, RL Envs and beyond that. ----------- Vivek Trivedy [Guest]: https://x.com/Vtrivedy10 Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://groundzeroai.in ------------ SPONSOR: Haize Labs (https://haizelabs.com) is building expert-level agents for mission-critical work, powered by proprietary Reliability Harness. They are doing some ambitious and interesting work. Checkout Leonard Tang (Cofounder and CEO) articles to learn more: https://x.com/leonardtang_/articles To sponsor future episodes, visit https://www.groundzeroai.in/partner ------------ TIMESTAMP: 0:00:00 - INTRO 0:03:30 - PhD at Temple and First Startup in Vision 0:12:42 - Bio Intern at Lockheed Martin and Mike Mill lore 0:16:18 - Building at Langchain, DeepAgents, R&D in Open Source 0:25:09 - Secret behind Harness Design, Filesystem as most Foundational Harness Primitive 0:42:25 - Trajectories for Continual Learning, Where and Where "NOT to" RL, Skills and Context Rot 0:56:30 - Harness Engineering: OpenClaw and Hermes, Why you might don't need a Claw, Custom Harnesses, Quick Fire 01:12:15 - Meta-Harness, Self-Improvement Loop, Simulation-as-a-Service, RL Envs (Computer-Use, Long-horizon) 01:28:30 - Advice to anyone 20yo / starting out college

944 views • Apr 22, 2026

1.52 hours

The Paradigm of Autonomous Research | Francesco Pappone (CEO, Paradigma)

Francesco Pappone (@tensorqt) is building Paradigma - an infrastructure for autonomous research. Their product Flywheel replaces papers with directed acyclic graphs where every experiment is a node with tracked lineage, and AI agents traverse and extend the graph via MCP. In this conversation, we go through Francesco's path from physics to founding Paradigma, how Flywheel works, building from Rome, how it connects to Karpathy's autoresearch and the broader auto-research movement, the hard problems around trust and verification when AI runs experiments, and where autonomous research is headed. ----------- Francesco Pappone [Guest]: https://x.com/tensorqt Himanshu [Host at GroundZero]: https://x.com/himanshustwts Try Flywheel: https://flywheel.paradigma.inc GroundZero: https://groundzeroai.in ------------ To sponsor future episodes, visit https://www.groundzeroai.in/partner ------------ TIMESTAMPS: 00:00:00 - INTRO 00:01:03 - Dropping out of PhD, Saturation in Academia, Building from EU 00:13:15 - Autoresearch, Idea behind Flywheel, ex-OpenAI Co-founder 00:19:18 - Evals for Autoreserch and Research Taste 00:25:00 - How a Research Lab uses Flywheel? 00:31:15 - Thesis and Autoresearch Adoption in Top 5 Labs 00:41:00 - Working around with Flywheel, Combinatorial Explosion 01:02:03 - Science as Public Graph, Deriving New Research Loop, Quick Fire 01:06:23 - Toughest Autoresearch Challenge as of Today, Future Directions 01:11:53 - Automating Science: Sakana and Deepmind, Trust Problem? 01:21:07 - Business Model, Automating R&D, Model Capabilities 01:27:20 - Dropout of college to live on frontier!

640 views • Apr 14, 2026

1.64 hours

Building Frontier Open Reasoning Models | Lucas and Varun (Arcee AI)

Arcee AI went from post training company to building their own frontier reasoning model from scratch + and open-sourcing all of it. In this episode, Lucas Atkins (Co-Founder & CTO) and Varun Singh (Pre-Training Lead) take us through the full story — the decision to start pre-training, how they built and trained Trinity on a fraction of the budget that frontier labs spend, the data and engineering challenges along the way, and their views on where open-source AI is headed. A really honest conversation about what it actually takes to build at the frontier. ----------- Lucas Atkins [Guest]: https://x.com/latkins Varun Singh [Guest]: https://x.com/stochasticchasm Himanshu [Host at GroundZero]: https://x.com/himanshustwts Try Trinity: https://chat.arcee.ai Weights (Apache 2.0): https://huggingface.co/arcee-ai GroundZero: https://groundzeroai.in ------------ SPONSOR: Huzzle Labs (https://labs.huzzle.com) is engineering and research company focused on RL environments - mainly code, computer use, and long-horizon enterprise workflows. Underneath that they have a separate talent marketplace with 300k experts and their own AI recruiter. They are already supporting frontier labs with RL data. So if you work in this space, research or human data, check them out! To sponsor future episodes, visit https://www.groundzeroai.in/partner ------------ TIMESTAMP: 00:00:00 - INTRO 00:00:59 - Varun's transition from SWE to Pre-Training Lead 00:04:20 - Trinity Manifesto, Openclaw Ecosystem 00:12:15 - Arcee's Post-Training to Pre-Training Pivot 00:23:45 - Varun's first Pre-Training Run (you can just do things!) 00:27:33 - Saturation in Pre-Training?, Mid-Training 00:37:00 - Tweaking the Training Architecture, Adam vs Muon, Evals 01:09:07 - Inference Engineering, Quick Fire, Post-Training Recipe 01:18:02 - Alpha in RL Envs, Harness Design 01:23:00 - American Open Source is trailing Chinese Competitors, Trinity Adoption 01:29:25 - Hiring at Arcee, Advice to 20yo

739 views • Apr 8, 2026

1.55 hours

Agents, Verifiability and First Principles Thinking | Vishnu Patankar (CTO, EigenCloud)

In this episode, Vishnu Patankar (CTO, EigenCloud) joins us for a wide-ranging conversation on AI, infrastructure, and the economy — from chips and compilers to agent harnesses, privacy/security, evals, deterministic inference, RL environments, open source, and where the frontier is headed. -------------------- Vishnu Patankar [Guest]: https://x.com/vishnu_patankar Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://groundzeroai.in EigenCloud: https://eigencloud.xyz --------------------- SPONSOR: HydraDB (https://hydradb.com) provides the developer tools and infrastructure to build, maintain, and scale an agent’s context and memory layer. This enables agents to persist context, interactions, and outcomes across sessions - making AI stateful. To sponsor future episodes, visit https://www.groundzeroai.in/partner -------------------- TIMESTAMPS: 00:00:00 - Intel Monopoly, Chips and Compilers 00:15:10 - OpenClaw, Guardrails and Harness 00:25:45 - First Principles of Writing Code 00:32:18 - Privacy and Security while running AI Systems 00:39:10 - Evals, Benchmarks and Taste 00:44:15 - Deterministic over Non-Deterministic Inference, Floating Point Maths 01:03:45 - Self-Improving Agents, RL Envs and Rollouts 01:15:10 - AGI, ASI and Continual Learning 01:20:50 - Open Source, Chinese Frontier 01:26:30 - K-Shaped Economy, Curiosity is the Frontier

552 views • Mar 28, 2026

1.52 hours

Tiny models will run the World | Vikhyat K (Co-founder, Moondream AI)

Bringing an amazing podcast with Vik, Co-founder of Moondream AI. We have discussed the importance of data quality, post training, and vision AI in particular. There are lot of easter eggs in this conversation. I hope you will enjoy this! More episodes: https://www.groundzeroai.in/podcasts SIGNALS: https://www.groundzeroai.in/signals COMMUNITY BLOGS with BOUNTIES: https://www.groundzeroai.in/spotlights -------------------------------------------------------------------- Vikhyat K [Guest]: https://x.com/vikhyatk Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://x.com/groundzero_twt ---------------------------------------------------------------------- To sponsor a future episode: https://www.groundzeroai.in/partner To partner or collaborate: ground0ai.lab@gmail.com ---------------------------------------------------------------------- TIMESTAMP: 00:00:00 - Vik's story and Moondream 00:18:50 - The core thesis of small and scale 00:28:35 - The data problem 00:33:35 - Deciding the training architecture 00:43:20 - Post-Training and RL perf 00:46:06 - Post-Training recipes of Moondream 3 00:47:40 - Open Source and VLM development priorities 00:52:05 - AI War : America and China 00:55:07 - Moondream acquisition and Future 01:04:08 - Community Questions 01:16:40 - Trivia, Lores and Opinions on current state of AI 01:28:07 - Advice to 20yo

1.2k views • Dec 16, 2025

0.99 hours

The Story of Dhravya Shah | 20yo raised $3M to build SuperMemory

Bringing an exciting pod with Dhravya Shah. We had talked about his journey and experiences. -------------------------------------------------------------------- Dhravya Shah [Guest]: https://x.com/DhravyaShah Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://x.com/groundzero_twt ---------------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts To partner or collaborate: ground0ai.lab@gmail.com ---------------------------------------------------------------------- TIMESTAMP: 00:00:00 - Teaser 00:01:39 - Introduction 00:02:42 - What is SuperMemory? Explaining the Product 00:04:43 - Coolest Use Cases & Customer Stories 00:07:48 - Early Days: Growing Up in Mumbai & Learning to Code 00:09:24 - First Success: Discord Bot & Twitter Screenshot Tool Acquisition 00:13:11 - The IIT Story: Myth vs Reality 00:14:38 - The 40-Week Building Streak 00:17:37 - Learning Strategy & Resources 00:20:18 - From AnyContext to SuperMemory: The Origin Story 00:21:16 - List of Failed Projects & Lessons Learned 00:25:04 - Getting Attacked & Accidentally Joining Cloudflare 00:26:30 - Relationship Support & Building While in College 00:27:57 - How to Sell Your Projects & Acquisitions 00:29:23 - Working at Mem0 & Key Differences with SuperMemory 00:33:51 - Cloudflare Experience & Working with CEO Dane Knecht 00:36:00 - The Fundraising Journey: From Buildspace to $3M Round 00:40:51 - Why Skip Y Combinator? 00:42:16 - O-1 Visa Story: Becoming "Officially Extraordinary" 00:44:14 - Being a Solo Founder: Challenges & Benefits 00:47:20 - Hiring Philosophy & Team Culture at SuperMemory 00:51:46 - India vs Bay Area: Ecosystem Differences 00:53:10 - Vision vs Profit: What Matters in Early Stage 00:54:26 - Thought on Joining College 00:55:18 - What's Next for SuperMemory (Local-First & Nova) 00:57:38 - Advice for Aspiring Builders & Students 00:59:00 - Closing Thoughts

5.8k views • Oct 12, 2025

2.08 hours

Model is the Product | Common Corpus, Mid-Training, Open Science

Bringing an exciting pod with Pierre-Carl Langlais (aka Alexandar Doria). We'd discussed about pre-training recipes, common corpus, mid-training, agentic systems, good post-training and everything AI. -------------------------------------------------------------------- Alexender Doria [Guest]: https://x.com/Dorialexander Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://x.com/groundzero_twt ---------------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts To partner or collaborate: ground0ai.lab@gmail.com ---------------------------------------------------------------------- TIMESTAMP: 00:00:00 - TEASER 00:01:12 - INTRO 00:02:03 - Who is Alexander Doria [Pierre-Carl Langlais]? 00:04:10 - Early career: From humanities to AI research 00:07:50 - Meeting influential people in computational humanities 00:10:00 - How the idea of Pleias came about 00:13:30 - Building Pleias: Infrastructure and compute challenges in Europe 00:17:06 - Team structure and work culture at Pleias 00:19:06 - What is "open science" and why it matters 00:21:53 - Big announcement: OpenSynthetic initiative 00:25:25 - Synthetic data experiments and surprising results 00:28:11 - "The Model is the Product" - explained 00:31:56 - Implications for companies building on top of models 00:35:25 - Differentiation in a world of shared base models 00:38:40 - Common Corpus: Origins and development 00:44:12 - The lack of open, legally clear datasets 00:47:03 - Anthropic's use of Common Corpus for mechanistic interpretability 00:50:20 - What makes good post-training? 00:54:00 - Reasoning under 400M parameters in SLMs 00:56:35 - Generalist scaling is stalling - where are the diminishing returns? 00:59:40 - Will specialization always win over scale? 01:02:00 - Opinionated and task-specialized models 01:06:29 - How inference cost drops change monetization models 01:09:12 - New value layers beyond token marketplaces 01:11:38 - Major technical obstacles to embedding workflows in models 01:13:40 - How smaller labs can compete on training infrastructure 01:15:36 - Should startups raise capital for AI training? 01:17:16 - What new capabilities do models need for orchestration? 01:19:50 - Designing verifier functions for agentic models 01:22:17 - RL in domains with weak or delayed rewards 01:24:50 - Multi-step training loops: Draft, verify, refine, backtrack 01:26:38 - The scarcity of agentic data and bootstrapping solutions 01:29:32 - Making agent training tractable at scale 01:31:44 - What is mid-training and why it matters 01:34:55 - Deployment, use cases, and hybrid model architectures 01:37:37 - Human-in-the-loop for regulated domains 01:39:48 - Advice for startups positioning in this transition 01:41:58 - Europe's structural challenges in AI 01:45:52 - Tokenizers: The overlooked competitive frontier 01:49:59 - Training LLMs on personal data and dead languages 01:52:12 - World models and JEPA architectures 01:53:50 - Building agentic systems: Stack and RL environments 01:55:34 - The art of training good RL models 01:58:49 - Trivia: Underrated habits and mindsets in research 02:00:09 - AI Twitter community and its impact 02:01:40 - Advice for folks starting in AI research 02:03:27 - Final thoughts and wrap-up

1.3k views • Oct 5, 2025

1.70 hours

The Lore of 20yo ML Researcher at Prime Intellect | RL, Agents and Intelligence

The Lore of Kalomaze! Bringing an exciting pod with Kalomaze (20 yo ml researcher, prime intellect). We'd discussed about training, fine-tuning, RL, scaling or llms in general and some interesting TPOT lores too. -------------------------------------------------------------------- Kalomaze [Guest]: https://x.com/kalomaze Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://x.com/groundzero_twt ---------------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts To partner or collaborate: ground0ai.lab@gmail.com ---------------------------------------------------------------------- TIMESTAMP: 00:00:00 - TEASER 00:01:06 - INTRO 00:01:49 - Organic Influence 00:02:46 - "Extremely Silly Jester" 00:04:04 - School Days, Diving into ML Research 00:09:27 - Developing "Research Taste" & The Art of Selection 00:11:45 - Dropping College, Parents' Reaction, Plan Ahead 00:18:13 - How a Reddit Post Led to a Job at Shopify 00:23:53 - Why He Chose Prime Intellect Over Other Offers 00:29:27 - A Day in the Life: Workload at PI 00:31:53 - Story of Verifiers, GRPO, and Semi-Verifiable Rewards 00:40:02 - Progress on Synthesizers 00:43:28 - The Environment Hub: A Hugging Face for RL? 00:47:47 - How ideas to research emerge, Experimentation 00:51:38 - Defining "Taste" in Research 00:54:15 - The Future of Supervised Fine-Tuning 01:01:48 - The Flaw in "Pure Reasoner" Models & Thoughts on GPT-OSS 01:05:34 - Perspective on the Chinese Open Source AI Surge 01:07:58 - The Art of Training a Good RL Model 01:10:08 - How to Learn a New Field: A Hacker's Approach 01:12:33 - Handling Disagreements with Formally Experienced Researchers 01:14:49 - Recipe to RLHF 01:19:47 - Scaling Bigger Models vs. Designing Better Rewards 01:26:31 - Rethinking Progress & The Post-AGI Narrative 01:29:27 - The Most Unexpected Part of Working at Prime Intellect 01:30:09 - Trivia Round Begins 01:30:25 - Code Reviews from Will? 01:33:43 - A Hidden Secret About Will! 01:34:26 - The Underrated Mindset That Gives Him an Edge 01:36:18 - Take on AI Tech Twitter ("TPOT") 01:38:16 - How Kalomaze shaped up + Advise 01:41:17 - Podcast Wrap-up

2.3k views • Sep 3, 2025

1.41 hours

LLMs for Everyone | Pre-training, Fine-Tuning, Scaling RL, Open Source | Daniel Han, Unsloth

Daniel Han on: * Early Career Trajectory * Founding Unsloth, GTM Strategy * LLM Pre-training, Fine-tuning, Scaling RL * How did YC help and much more.. -------------------------------------------------------------------- Daniel Han [Guest]: https://x.com/danielhanchen Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero: https://x.com/groundzero_twt ---------------------------------------------------------------------- This episode is powered by Bhindi AI (cursor for your apps): https://bhindi.io ---------------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts To partner or collaborate: ground0ai.lab@gmail.com ---------------------------------------------------------------------- TIMESTAMP (00:00:00) - TEASER (00:01:00) - INTRO (00:02:10) - Early Career (00:04:47) - Hyperlearn and Sciblox (00:09:30) - Bug Bounty Guy (00:12:32) - Questions for Daniel (00:30:20) - Scaling RL, GRPO and Transformer moment (00:42:35) - Unsloth: GTM, YC aid, next big thing and more (00:59:10)- General: learning process, multi-gpu, advices and more

5.6k views • Aug 16, 2025

1.26 hours

Soham Parekh Unplugged with Himanshu | GroundZero YT

Welcome to another episode of GroundZero Talks. This is our first in-person episode. Today, we have Soham Parekh or "Everyone's favorite founding engineer". Early July, X users have shared dozens of stories about encounters with Soham Parekh, a software engineer who seems to have been simultaneously working at multiple Silicon Valley startups. We'd discussed his whole journey, experiences and reflections in this episode. Hope you'll find it insightful. ---------------------------------------------------------------- Soham Parekh [Guest]: https://x.com/realsohamparekh Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero AI: https://x.com/groundzero_twt Update: We ended up launching a company together (https://physera.ai/) ---------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts To partner or collaborate: himanshu@groundzeroai.in ------------------------------------------------------------------- TIMESTAMPS (00:00:00) - TEASER (00:00:45) - INTRO (00:02:40) - Define Soham Parekh, Early Career (00:05:10) - Current routine and Interests (00:06:38) - Soham - a Generalist? (00:08:05) - How has been life since twitter blowout? (00:09:17) - Academics, Dropping out, Grad School (00:10:38) - Acing Tech Interviews, Open Source (00:11:44) - Curious about Low-level internals (00:14:30) - Mistakes, Tutorial hell, Interviews (00:17:12) - What was the first immediate reaction? (00:18:45) - Reaction of family and friends (00:20:10) - What were the misconceptions floating around? (00:21:15) - Dev-shop of 20 Interns?, Interviews, Contributions (00:25:03) - Did Soham give any company his 100%, working parallely? (00:27:55) - How picky Soham was choosing startups to interview? (00:29:58) - How was a typical day with peak workload? (00:31:20) - Why should anyone trust your words/work ethic? (00:32:32) - Incurred interest on integrity, Work Trials (00:35:10) - Crazy story with a startup (00:36:25) - How much Soham used Cursor/AI tools with jobs? (00:38:25) - Did Soham interview at Frontier AI Labs? (00:40:02) - Sharing MVPs among different companies? (00:42:38) - Mutual understanding/friendships with founders? (00:44:38) - Managing onboarding at multiple startups? (00:48:12) - Surviving background checks (00:49:56) - Difference between culture in SF and India (00:52:08) - Thoughts on Agentic AI space (00:58:13) - Which one startup excited Soham most? (01:01:00) - Contradictory statement in TBPN Interview, self perception now (01:02:14) - Negative stereotype reinforced for remote work (01:04:45) - What is one thing startups are completely missing now? (01:07:15) - What kind of tech product Soham will build in current era? (01:09:30) - How would Soham react with moonlighting as a CEO? (01:12:05) - What excites Soham more with current startup (Darwin) (01:14:05) - Final thoughts (01:15:21) - CHEERS!

19.3k views • Jul 19, 2025

1.72 hours

RL, Reasoning, Reward Hacking, AI Timeline and Post AGI | Will Brown (Research at Prime Intellect)

Will Brown on: - Early Career Trajectory (Industry and Academia) - GenAI Handbook - RL and Reasoning - Self-Improving Agents - AI Timeline - Reward Hacking - Post AGI Landscape ------------------------------------------------------------- Will Brown [Guest]: https://x.com/willccbb Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero AI: https://x.com/groundzero_ai ------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts To partner or collaborate: groundzeroai.lab@gmail.com --------------------------------------------------------------- TIMESTAMPS (00:00:00) - INTRO (00:02:00) - Career Trajectory, Love for Maths and Algorithms (00:06:24) - Early Motivation and PhD experience (00:12:22) - Teaching, Open Source, GenAI Handbook (00:17:27) - Rubrik Engineering and Approach to Design (00:23:06) - RL over Instruction Tuning (00:26:07) - Journey behind VERIFIERS (00:33:30) - RL experiments on SLMs (00:37:06) - Decentralized Training and Inference (00:40:28) - Self-Improving agents (00:44:05) - GRPO in Multimodal settings, Robotics (00:49:00) - Challenges in Curriculum Learning (00:52:02) - Kimi Dev 72B? (00:54:55) - Reinforcement Pre-training (00:57:53) - Do LLMs have hit a wall (in novelties)? (01:02:06) - LLM and Classical Agents (01:05:10) - LLMs as Evaluators or Doers (01:07:19) - Do models know when they're Reward Hacking? (01:12:42) - Training agents in diverse environments (01:14:55) - What if solving reward hacking accidentally gives us alignment? (01:16:28) - Will AGI collapse decentralized system into centralized one? (01:19:24) - The Illusion of the Illusion of Thinking (01:20:21) - LLMs in Quant Trading (01:22:21) - Thoughts on Pattern Matching in Intelligence? huhhh, let's relax! (01:23:55) - Kalomaze's hidden secret (01:24:30) - Which frontier lab Will is bullish on? (01:26:57) - Finest teams in AI Research (01:30:45) - Is xAI late compared to other frontier labs? (01:32:35) - How to become an Experimentalist? (01:33:49) - Underrated mindset that has been an edge in Will's research? (01:34:41) - "Cheat Code" in RL Research that has been game changing? (01:35:45) - What Will think about AGI? (01:38:01) - Will's advice to 20yo high school/college grads (01:39:23) - Classical ML in current age of AI Research (01:41:10) - What Will have to say for a general audience? (01:42:29) - Concluding!

4.2k views • Jun 22, 2025

2.32 hours

Post-Training, RL, Experiments and Indic AI | Tokenbender

Tokenbender on: - Post Training - RL and Reasoning - Post AGI Landscape - Experiments - Indic AI Landscape ------------------------------------------------------------- Tokenbender [Guest]: https://x.com/tokenbender Himanshu [Host at GroundZero]: https://x.com/himanshustwts GroundZero AI: https://x.com/groundzero_ai -------------------------------------------------------------- To sponsor a future episode: https://buymeacoffee.com/himanshustwts --------------------------------------------------------------- TIMESTAMPS (00:00:00) - INTRO (00:01:30) - Career Trajectory and Motivation (00:10:00) - Non-CS Background and Building Intuitions (00:14:00) - Journey with Codecherrypop and Small Models (00:21:50) - Partner-In-Crime and Roleplay Series (00:28:45) - Post-Training and how it is evolved? (00:36:45) - Is pre-training actually dead? (00:45:10) - RL over next-token-predictors? (00:52:10) - Reliable agents, RL in training workload (00:58:07) - Weak priors and Reward Sparsity (01:01:20) - What's new RL sauce? (01:06:11) - RL from Zero Pre-train, Coherent text and Beyond (01:12:01) - Intelligence isn't flat, Optimizing for one sharp spike? (01:16:37) - Sampling and Creating Data for Models, New approaches? (01:20:55) - Role of failures (01:24:26) - Obsession over next number (01:27:05) - Shallow safety alignment (01:31:58) - RL over Diffusion Models, 'aha' moments (01:37:40) - 50x in productivity? (01:40:18) - How do you build the mindset to keep experimenting? (01:48:20) - Writing papers on AI research (01:51:10) - How you look up to open source models, what next? (01:53:45) - Finding or Creating synthetic datasets (01:56:08) - TRIVIA (02:07:06) - Indic AI Landscape, Challenges (02:12:32) - ADVICE FOR STUDENTS (02:16:50) - FINAL THOUGHTS FROM TOKENBENDER

3.6k views • Jun 1, 2025

1.51 hours

ML Research | Career in Japan | AI4Bharat | GroundZero AI Talks EP01 with Raj Dabre

Welcome to GroundZero AI Talks EP01 with Raj Dabre! Agenda: Machine Learning Unplugged: Research, Trends, and the Road Ahead Dr. Raj Dabre is a: - Researcher- NICT, Japan - Adjunct Faculty - IIT Madras (AI4Bharat) - Visiting Professor- IIT Bombay

2.1k views • Jan 12, 2025

Last updated: Jun 1, 2026