Unsupervised Learning: With Jacob Effron
Yann LeCun on What Comes After LLMs
updated
The conversation covers everything from moving drug discovery out of the lab and into computers, to why AI models think differently than human chemists, to the strategic decisions around open sourcing foundational models while keeping design capabilities proprietary. It's an in-depth look at how AI is fundamentally changing pharmaceutical innovation and what it means for the future of medicine.
Check out the full Chai-2 Zero-Shot Antibody report linked here:
biorxiv.org/content/10.1101/2025.07.05.663018v1.full.pdf
0:00 Intro
2:10 The Evolution of AI in Drug Discovery
6:09 Current State and Future of AI in Biotech
11:15 Challenges and Modalities in Therapeutics
15:19 Data Generation and Model Training
23:59 Open Source and Model Development at Chai
28:35 Protein Structure Prediction and Diffusion Models
30:57 Open Source Models and Their Impact
35:41 How Should Chai-2 Be Used?
39:34 The Future of AI in Pharma and Biotech
43:51 Key Milestones and Metrics in AI-Driven Drug Discovery
48:24 Critiques and Hesitation
55:06 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
In this episode, Simon Eskildsen, co-founder and CEO of TurboPuffer, lays out a compelling vision for how AI-native infrastructure needs to evolve in an era where every application wants to connect massive amounts of context to large language models. He breaks down why traditional databases and even large context windows fall short—especially at scale—and why object-storage-native search is the inevitable next step. Drawing on his experience from Shopify and Readwise, Simon introduces the SCRAP framework to explain the limits of context stuffing and makes a clear case for why cost, recall, performance, and access control drive the need for smarter retrieval systems. From practical lessons in building highly reliable infra to hard technical problems in vector indexing, this conversation distills the future of AI infra into first principles—with clarity and depth.
0:00 Intro
0:49 The Evolution of AI Context Windows
2:32 Challenges in AI Data Integration
3:56 SCRAP: Scale, Cost, Recall, ACLs, and Performance
9:21 The Rise of Object-Oriented Storage
16:47 Turbo Puffer Use Cases
22:32 Challenges in Vector Search
27:02 Challenges in Query Planning and Data Filtering
27:53 Focusing on Core Problems and Simplicity
28:28 Customer Feedback and Future Directions
29:11 Reliability and Simplicity in Design
30:39 Evaluating Embedding Models and Search Performance
32:17 The Role of Vectors in Search Engines
34:16 Balancing Focus and Expansion
35:57 AI Infrastructure and Market Trends
38:36 The Future of Memory in AI
43:01 Table Stakes for AI in SaaS Applications
45:55 Multimodal Data and Market Observations
46:57 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
The conversation covers everything from folding laundry with robots to building scalable data pipelines, the limits of simulation, and what it’ll take to bring robot assistants into everyday homes. It's a wide-ranging and thoughtful look at where robotics is headed, as well as how fast we might get there.
Note: Redpoint was fortunate to participate in a fundraising round for Physical Intelligence in November 2024.
0:00 Intro
1:31 Early Days in Robotics
2:08 Shift to Learning-Based Robotics
4:50 Challenges and Breakthroughs
8:45 Google's Role and Spin-Out Decision
14:27 Comparing Robotics to Self-Driving Cars
18:37 Hardware and Intelligence
20:24 Future Milestones and Scaling Challenges
32:42 Data Collection and Infrastructure Needs
35:08 Choosing and Tackling Complex Tasks
38:08 Evaluating Model Performance
40:47 The Role of Simulation in Robotics
43:46 Research Strategies and Hiring
47:35 Open Source and Community Impact
51:46 Advancements in Training and Model Efficiency
58:04 Future of Robotics and AI
1:00:35 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:49 Launching a New Startup: LMArena
1:01 The Origin of the Vicuna Model
1:54 Challenges in Model Evaluation
6:33 Becoming a Company
7:47 Expanding Evaluation Capabilities
13:48 The Importance of Human-Based Evaluations
18:56 Open Source vs. Proprietary Models
23:05 Infrastructure and Collaboration Challenges
28:22 China's Strategic Advantages in Technology
29:54 Opportunities in AI Infrastructure
31:50 Challenges in AI Model Optimization
35:49 The Role of Data in AI Enterprises
39:31 Reflections on AI Progress and Predictions
50:40 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
They’re operating at the intersection of recruiting, evals, and foundation model development—helping companies shift from intuition to measurable prediction. Brendan and his team recently raised $100M and are working with some of the most advanced players in the AI ecosystem today.
0:00 Intro
1:29 State of AI in Talent Evaluation
2:06 Improvements in AI Models
4:19 Mercor Background and Mission
5:21 AI Use Cases in Hiring
13:55 Data Labeling Landscape
17:00 Expanding Beyond Coding
18:51 Company Vision and Market Strategy
21:23 Meeting with xAI
23:59 Does Mercor Use Their Own Product?
25:53 Exploring Multimodal Capabilities
28:15 Skills for the Future: Embracing AI
29:41 The Demand for Software Engineers
35:07 Foundation Model Landscape
38:54 AI Regulations
40:09 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
1:30 The Evolution of AI in Law
2:43 AI's Impact on Legal Processes
8:28 Advantages Over Other Players in the AI Law Space
12:19 Challenges in Educating Users
17:28 The Hardest Part of Building Legora
18:46 Pricing Models and Cost Management
25:42 YC Experience and Commercial Focus
28:11 Being Patient When Releasing Products
30:58 Maintaining a Fast-Paced Work Culture
33:24 Rapid Growth and Market Penetration
36:59 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:48 Claude 4
1:30 Capabilities and Improvements
2:29 Practical Applications and Advice
3:04 Future of AI in Coding
4:38 Managing Multiple AI Models
11:05 The Barrier to Agents is Reliability
16:20 Agents Conducting Research
19:39 Impact of Models on World GDP
24:59 Most Important Metrics in Model Improvement
29:38 Stories of Model Creativity
32:30 How Often Will New Models Be Shipped in the Future?
39:36 Day-to-Day Work of AI Researchers
46:31 The Future of AI and Society
51:11 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Authors @Daniel Kokotajlo (former OpenAI researcher now focused full-time on alignment through his nonprofit, @AI Futures, and one of TIME’s 100 most influential people in AI) and @Thomas Larsen joined the show to unpack their findings.
We talk through the key takeaways from the report, its policy implications, and what they believe it will take to build safer, more aligned models.
0:00 Intro
1:08 Overview of AI 2027
2:25 AI Development Timeline
4:03 Race and Slowdown Branches
12:45 US vs China
18:02 Potential AI Misalignment
31:05 Getting Serious About the Threat of AI
47:22 Predictions for AI Development by 2027
48:32 Public and Government Reactions to AI Concerns
49:26 Policy Recommendations for AI Safety
52:21 Diverging Views on AI Alignment Timelines
1:01:29 The Role of Public Awareness in AI Safety
1:02:37 Reflections on Insider vs. Outsider Strategies
1:10:52 Future Research and Scenario Planning
1:14:00 Best and Worst Case Outcomes for AI
1:17:01 Final Thoughts and Hopes for the Future
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
If you’ve ever wondered how OpenAI really decides what to build, and how it affects what you should build, this one’s for you.
0:00 Intro
0:50 Deep Dive into GPT-4.1 Development
2:10 User Feedback and Model Evaluation
3:48 Challenges and Improvements in Model Training
5:41 Advancements in AI Coding Capabilities
8:58 Future of AI Models and Fine-Tuning
20:31 Multimodal Capabilities
22:46 Deep Tech Applications and Data Efficiency
24:01 Preference Fine Tuning vs. RFT
26:16 Choosing the Right Model for Your Needs
28:05 Prompting Techniques and Model Improvements
31:57 Future Research and Model Enhancements
39:01 Power Users and Personalization
40:09 Personal Journey and Organizational Growth
43:24 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
🎧 Sal joined us on Unsupervised Learning to talk about AI's role in education— from the vantage point of someone deploying it at scale. As founder of Khan Academy, he’s overseen the rollout of AI tools to over a million teachers and students, giving him a front-row seat to what’s actually working in classrooms.
0:00 Intro
1:40 The Vision for Future Classrooms
4:51 Khan Academy's AI Initiatives
7:29 Building Proactive AI
10:57 Teacher and Student Experiences
18:47 District-Level Adoption and Policy
26:19 Gamification and Student Engagement
27:59 Evaluating AI Models
31:26 Global Impact and Future Prospects
38:06 Challenges and Innovations in AI Education
44:42 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
If you’re building or investing in AI infrastructure, Aidan is worth listening to. He co-authored the original Transformer paper, leads one of the most advanced model labs outside of the hyperscalers, and is now building for real-world enterprise deployment with Cohere’s agent platform, North. Cohere serves thousands of customers across sectors like finance, telco, and healthcare — and they’ve made a name for themselves by staying model-agnostic, privacy-forward, and deeply international (with major bets in Japan and Korea)
0:00 Intro
0:32 Enterprise AI
3:23 Custom Integrations and Future of AI Agents
4:33 Enterprise Use Cases for Gen AI
7:02 The Importance of Reasoning in AI Models
10:38 Custom Models and Synthetic Data
17:48 Cohere's Approach to AI Applications
23:24 Future Use Cases and Market Fit
27:11 Building a Unified Automation Platform
27:34 Strategic Decisions in the AI Journey
29:19 International Partnerships and Language Models
31:05 Future of Foundation Models
32:27 AI in Specialized Domains
34:40 Challenges in Data Integration
35:06 Emerging Foundation Model Companies
35:31 Technological Frontiers and Architectures
37:29 Scaling Hypothesis and Model Capabilities
42:26 AI Research Culture and Team Building
44:39 Future of AI and Societal Impact
48:31 Addressing AI Risks
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:39 AI Investment Landscape
1:30 Market Projections
3:25 Strategic Imperatives in AI
6:04 AI Model Layer Insights
10:06 Infrastructure Layer
11:46 Application Layer
15:04 Vertical AI SaaS Opportunities
21:45 Evaluating Early-Stage Founders
23:39 First Mover Advantage and Competitive Dynamics
25:28 Domain Expertise vs. AI Expertise
26:58 Why AI Startups Fail
29:41 Startups vs. Incumbents
35:15 Navigating High AI Valuations
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Swyx brings deep experience from his time at AWS, Temporal, and Airbyte, and is now focused on AI agents and dev tools. Alessio is an investor at Decibel, where he’s been backing early technical teams pushing the boundaries of infrastructure and applied AI. Together they run Latent Space, a technical newsletter and podcast by and for AI engineers
To subscribe or learn more about Latent Space, click here: youtube.com/@LatentSpacePod
0:00 Intro
1:08 Reflecting on AI Surprises of the Past Year
2:24 Open Source Models and Their Adoption
6:48 The Rise of GPT Wrappers
7:49 Challenges in AI Model Training
10:33 Over-hyped and Under-hyped AI Trends
24:00 The Future of AI Product Market Fit
30:27 Google's Momentum and Customer Support Insights
33:16 Emerging AI Applications and Market Trends
35:13 Challenges and Opportunities in AI Development
39:02 Defensibility in AI Applications
42:42 Infrastructure and Security in AI
50:04 Future of AI and Unanswered Questions
55:34 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:53 OpenAI’s Vision for Consumer Interaction
4:51 Building Multi-Agent Systems for Business Solutions
6:53 Challenges and Innovations in AI Fine-Tuning
13:20 Exploring Computer Use Cases and Applications
17:20 Advanced Use Cases and Developer Insights
25:29 Challenges with Context Storage and Chat Completions
26:09 Introducing the Responses API and MCP
27:16 AI Infrastructure Companies and Their Role
29:35 Building the Tools Ecosystem
30:17 Exploring Computer Use Models
31:47 The Future of AI and Developer Tools
38:36 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
🔹 Noam Shazeer, co-inventor of the Transformer
🔹 Jack Rae, Research Director at DeepMind and one of the leads behind Gemini’s Flash Thinking
We got to ask them all of the top-of-mind questions in AI today about where we are, where we’re headed and what it means for businesses and the world. Some key take-aways:
0:00 Intro
2:07 Exploring Gemini 2.0
4:41 Challenges with Evals and Benchmarks
6:51 Reinforcement Loops and AI Productivity
8:52 Agentic Coding and AI in Development
13:39 Multimodal Capabilities and Applications
15:58 Future of AI: Complexity and Reliability
19:39 Test Time Compute and Data Efficiency
31:57 AI Research Culture and Breakthroughs
39:36 Reflecting on Large Language Models
40:14 The Rise of Vision-Based Models
41:38 Native Image Generation and General Purpose Models
42:12 AI in Healthcare and Specialized Models
44:09 Shifting Timelines and Rapid Adoption
47:25 Open Source Models and Competitive Edge
50:07 AI's Impact on Education and Personal Lives
55:47 AGI Risks and Safety Considerations
58:10 Future of AI Companions
1:02:54 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
1:16 AI's Role in Energy and Climate Change
5:05 Innovative Energy Solutions
15:23 Open Source and AI Development
23:08 Challenges in Chip Design
24:37 Balancing Data Center Capacity
26:28 The Future of VR and AI Integration
30:14 AI's Role in Climate Solutions
32:14 AI in Material Science and Beyond
35:04 Personal AI Assistants and Their Impact
39:20 Reflections on AI and Future Predictions
41:56 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
In this week’s episode of Unsupervised Learning, we dive deep into the frontier where AI meets hardware — and there’s no better guide than Vincent Vanhoucke, Distinguished Engineer at Waymo and former Head of Robotics at DeepMind.
0:00 Intro
1:29 Waymo's Technological Evolution
3:19 The Role of LLMs in Autonomous Driving
6:41 Vincent's Journey to Waymo
9:56 Challenges in Autonomous Driving
12:37 Simulation and World Models
28:23 Future Milestones and Expansion
30:49 Broader Robotics and AI
36:51 Future of General Robotics Models
38:53 Hardware vs. Software Approaches in Robotics
40:58 Challenges in Robotic Data Acquisition
41:17 Simulation vs. Real-World Data in Robotics
43:41 Human-Robot Interaction for Data Collection
45:42 Advancements in Multimodal Models
47:47 Unanswered Questions in Robotics
52:41 Reasoning Capabilities in AI
55:36 Future of Robotics and AI
1:01:30 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
In this episode we focused on how far test-time compute gets us, the real implications of DeepSeek, what agents milestones he’s looking for and more.
0:00 Intro
1:37 DeepSeek Reactions and Market Implications
4:23 Challenges in Building AGI
8:22 Research Problems in AI Development
11:38 The Future of AI Agents
15:36 How to Make a Large Action Model
20:08 The Path to Reliable Agents
24:29 Future Human-Computer Interaction
25:23 Specialized Models and Policy
28:59 Amazon's Role in AGI Development
30:56 Data Labeling and Team Building
37:00 Reflections on OpenAI
42:35 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:47 Reasoning Models and Their Uneven Progress
2:47 Challenges in AI Benchmarks and Real-World Applications
5:04 Inference Scaling and Verifier Imperfections
7:34 Agentic AI: Tools vs. Autonomous Actions
12:08 Future of AI in Everyday Life
15:35 Evaluating AI Agents and Collaboration
24:50 Regulatory and Policy Implications of AI
27:50 Analyzing Generative AI Adoption Rates
29:18 Educational Policies and Generative AI
30:10 Flaws in Predictive AI Models
31:32 Regulation and Safety in AI
33:48 Academia's Role in AI Development
36:14 AI in Scientific Research
38:23 AI and Human Minds
46:05 Economic Impacts of AI
49:43 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Follow SemiAnalysis: semianalysis.com
0:00 Intro
1:27 Grading the AI Diffusion Rule
4:08 What Will Happen to the Malaysian Data Centers?
7:43 How do the Regulations Favor Giant Tech Companies?
9:27 Pre-Regulation AI Landscape
13:16 Where Does Chinese AI Go From Here?
22:17 The Goldie Locks Approach to Regulation
24:36 Size of Cluster Buildouts Today
38:07 How Big Will Cluster Buildouts Get?
43:20 Are Open-Source Models Falling Behind?
48:11 Questions Dylan Wants the Answers To
51:50 Hardware Startups
1:01:25 The Future of Enterprise AI
1:05:30 What Made CoreWeave So Successful?
1:19:48 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
He’s seen it all and we were fortunate to get his insights and vision for the future in one of my favorite episodes of Unsupervised Learning to date:
0:00 Intro
0:44 Debating AI Model Capabilities
0:57 Inside vs Outside Perspectives on AI Progress
1:39 Challenges in AI Pre-Training
3:02 Reinforcement Learning and Future Models
3:48 AI Progress in 2025
5:58 New Form Factors for AI Models
8:56 Reliability and Enterprise Integration
18:14 Multimodal AI and Video Models
24:05 The Future of Robotics
32:46 The Complexity of Automating Jobs with AI
34:08 AI in Startups: Tackling Boring Problems
35:33 AI's Impact on Productivity and Consultants
36:43 Traits of Top AI Researchers
40:52 The Evolution of OpenAI's Mission
46:57 The Challenges of Scaling AI
49:16 The Future of AI and Human Agency
54:47 AI in Social Sciences and Academia
1:01:15 Reflections and Future Plans
1:02:57 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
1:06 Fireworks: Revolutionizing AI Inference
2:12 Challenges in AI Model Development
4:05 The Future of AI: Compound Systems
4:32 Designing Effective AI Tools
10:26 Customization and Fine-Tuning in AI
14:06 Human-in-the-Loop Automation
16:38 Evaluating AI Models
19:18 Building Complex AI Systems
21:18 Function Calling and AI Orchestration
26:52 AI Infrastructure and Hardware
31:08 Small Expert Models
31:27 Hyperscalers and Resource Management
32:14 Inference Systems and Scalability
33:08 Running Models Locally: Cost and Privacy
35:20 Open Source Models and Meta's Role
36:41 The Evolution of AI Training and Inference
38:04 Fireworks' Vision and Market Strategy
40:46 The Impact of Generative AI
45:18 AI Research and Future Trends
46:58 Building for a Rapidly Changing AI Landscape
49:36 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
We discuss the technical breakthroughs that set o1 apart, its unique capabilities compared to previous models, and how it disrupts traditional paradigms in AI development. Noam also shares insights into OpenAI’s approach to innovation, the economic realities of scaling AI, and what the future holds for the field.
0:00 Intro
1:16 Scaling Model Capabilities and Economic Constraints
3:14 Excitement Around Test Time Compute
5:16 Challenges and Future Directions in AI Research
8:37 Noam Brown's Journey and OpenAI's Research Focus
16:34 The Role of Specialized Models and Tools
22:04 Unexpected Use Cases and Future Milestones
24:30 Proof of Concept: o1's Capabilities
25:34 The Bitter Lesson: Insights from Richard Sutton
26:45 Scaffolding Techniques and Their Future
28:42 Challenges in Academia and AI Research
31:16 Evaluating AI Models: Metrics and Trends
35:33 The Role of AI in Social Sciences
40:37 AI Agents and Emergent Communication
41:12 Future of AI Robotics
42:08 Advancing Scientific Research with AI
44:27 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Russ talked about when voice makes sense as an interface, the exciting new UX paradigms on the horizon, the intersection of voice and robotics and Anthropic's Computer Use API.
0:00 Intro
0:41 Using ChatGPT Voice in Daily Life
02:43 How LiveKit Works with ChatGPT Voice
05:33 LiveKit as the Nervous System for AI
08:28 Future of Work and AI Interfaces
18:48 Emerging Use Cases for Voice AI
22:46 AI Models in Customer Support
23:27 Latency Improvements in AI
24:54 Challenges in System Integration
26:18 Multimodal AI and Browser Integration
29:57 Telephony and AI in Healthcare
32:28 Humanoid Robotics and On-Device AI
34:07 Cloud vs. On-Device Inference
37:07 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:51 Incentives and Team Motivation at Databricks
02:39 The Evolution of AI Models: Transformers vs. LSTMs
5:26 Mosaic and Databricks: A Strategic Merger
7:30 Guidance on AI Model Training and Fine-Tuning
11:10 Building Effective AI Evaluations
16:01 Domain-Specific AI Models and Their Importance
19:36 The Future of AI: Challenges and Opportunities
25:36 Ethical Considerations and Human-AI Interaction
29:18 Challenges in AI Explainability
29:30 Databricks' Approach to AI Integration
29:42 Customer Collaboration and AI Implementation
31:14 Navigating AI Tools and Techniques
36:10 The Role of Open Source Models
37:15 AI Infrastructure and Partnerships
48:56 Academia's Role in AI Research
52:38 Ethics and Policy in AI
58:18 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:38 The Rise of AI and DeepL's Journey
1:41 DeepL's Competitive Edge and Market Impact
2:41 Innovations in AI Translation
4:39 DeepL's Product and Use Cases
7:35 Challenges and Strategies in AI Translation
14:29 Human Translators and Data Labeling
24:39 Building and Scaling AI Infrastructure
32:52 Evaluating AI Models: Objective vs Subjective
34:05 Translation Quality Metrics
35:35 The Debate on AI Moats and Specialized Models
40:16 The Impact of Real-Time Translation on Business
45:05 Challenges in Developing Synchronous Speaking Models
46:49 Adjacent AI Technologies and Their Potential
48:33 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:37 HeyGen's Viral Moments
1:23 Creating Magic with AI
3:30 The Future of AI in Video Production
9:29 HeyGen's Use Cases and Customer Base
13:15 AI Avatars
25:46 The Future of Content
26:43 Competing with Industry Giants
27:16 Innovating for New Markets
31:24 Enterprise Push: Lessons and Surprises
33:07 Trust and Safety in AI
37:03 Fundraising and Financial Strategies for AI Startups
41:22 Quickfire
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Percy is also a trained classical pianist, which clearly influences the way he thinks about technology. We explored the evolution of AI from simple token prediction to autonomous agents capable of long-term problem-solving, the problem of interpretability, and the future of AI safety in complex, real-world systems.
0:00 Intro
0:40 Percy Liang's Insights on AI Models
3:20 Challenges and Innovations in AI Benchmarks
9:59 The Role of Academia in AI Research
15:09 AI Safety and Regulatory Perspectives
22:09 Exploring Generative Agents and Simulations
32:38 Exploring Evaluation Methods
34:22 Vertical-Specific Benchmarks
35:43 Challenges in Interpretability
39:50 The Future of Model Architectures
43:45 Inference Market Innovations
50:10 Robotics and AI: The Next Frontier
52:51 AI in Music: A New Era
56:18 AI as Educators and Coaches
56:57 Quickfire
59:42 Jacob and Pat Debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
0:52 Exploring the Impact of Systems Thinking in AI
1:50 Latency Constraints and AI Deployments
2:06 Benchmarks and Real-World Applications
3:28 Transition to Contextual and Company Vision
5:13 Challenges and Innovations in Enterprise AI
8:52 The Evolution and Future of RAG
15:27 Alignment and Reinforcement Learning in AI
23:53 Collaborations and the Role of Academia
29:16 The Evolving Role of AI Developers
30:20 Changing Perspectives in AI Research
30:45 Synthetic Data and Agentic Workflows
33:48 The Future of Multimodal Data
35:32 Reasoning Capabilities in AI Models
42:57 The Rise of Multi-Agent Systems
45:25 Hugging Face and the AI Ecosystem
47:00 Building Contextual and AI Startups
49:52 The Future of AI and Personalized Entertainment
50:42 Quickfire Round
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Grammarly CEO Rahul Roy-Chowdhury believes AI will enable everyone to focus on more meaningful, creative interactions by automating the "drudgery" of daily tasks. It was interesting to hear him talk about how he thinks about competition and his longer-term perspective on how AI will be adopted by the enterprise.
0:00 Intro
1:08 The Future of AI in Human Communication
3:52 Grammarly's Evolution and Product Overview
8:26 Limitations of LLMs
15:36 The Impact of ChatGPT and Future Prospects
23:57 Fine-Tuning AI for User Needs
30:21 Competitive Landscape and Differentiators
39:19 AI in Education
46:30 Over-hyped/Under-hyped
50:02 Most Exciting AI Startups
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
That’s why it was so fascinating to speak with Baris Gultekin, Snowflake’s Head of AI, on our latest Unsupervised Learning. Baris has helped Snowflake launch several key products, including Cortex, AI Data Cloud, and even Snowflake’s own LLM, called Arctic. Baris was a founder himself — he joined Snowflake through an acquisition of his blockchain API startup, nxyz. He has a unique window into the future of AI via his role building key infrastructure. It was really fun to dive deep into this side of the AI ecosystem with Baris.
0:00 Intro
0:33 Snowflake's AI Product Portfolio
0:52 Building Arctic LLM: Challenges and Innovations
2:57 Use Cases and Applications of Arctic LLM
3:10 Cortex: Snowflake's Managed Service for LLMs
4:26 Tackling BI Challenges with Cortex Analyst
8:46 Data Extraction and Analysis with AI
10:15 Snowflake's AI Strategy and Future Directions
13:12 Governance and Security in AI Deployments
22:25 Supporting and Integrating New AI Models
24:02 Discussing the 405B AI Model
25:02 Enterprise Use Cases and Cost Considerations
26:09 Future of Arctic LM and Security Concerns
26:59 Open Source Models and Guardrails
30:21 Internal Use of LLMs at Snowflake
31:32 Comparing Snowflake and Databricks AI Strategies
33:24 Opportunities for AI Startups
34:14 Exciting AI Developments and Future Prospects
40:34 Over-hyped/Under-hyped
45:58 Closing Thoughts and Resources
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
We learned so much about how language works and how Connor has built this startup. Some of our favorite bits:
0:00 Intro
0:53 Connor's Entrepreneurial Journey
3:55 Diving into AI and Language Learning
6:22 The Evolution of Speak
9:45 Building Specialized Models and Overcoming Challenges
18:32 User Experience and Interface Design
24:33 Future of AI in Language Learning
35:53 Comparing Duolingo and Speak
38:15 Challenges in Translation and Human Connection
41:33 Specialized AI Models and Their Impact
47:56 Opportunities in Professional and Personal Learning
53:53 The Evolution of Education with AI
59:46 Final Thoughts and Reflections
1:03:22 Jacob and Pat Debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 Intro
1:10 The Magic of GitHub Copilot
2:09 AI's Impact on Software Development
2:51 Global Adoption and Democratization of AI
4:22 Keeping Developers in the Creative Flow
7:01 Future of Software Development with AI
9:56 Challenges and Opportunities in AI-Powered Coding
11:33 The Role of Agents in Copilot's Strategy
17:05 AI's Influence on Open Source Ecosystem
24:10 Fine-Tuning and Customization of Copilot
28:24 The Rapid Evolution of Copilot
30:28 Future Innovations and Accelerating Pace of AI
33:11 The Future of AI in Software Development
33:46 AI's Impact on Different Tech Stacks
34:23 The Evolution of Media Consumption
35:17 The Competitive Landscape of AI Models
37:04 The Infinite Game of Business
38:19 The Role of Multiple AI Models in Enterprises
43:03 Advice for Founders Competing with Incumbents
45:08 The Importance of Focus in Startups
45:52 Expanding the GitHub Ecosystem with Extensions
47:22 The Next Wave of Copilot
50:59 Quickfire
54:37 Erica and Jordan Debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
(0:00) intro
(1:07) how early are we in the AI for creative tools space
(3:31) how Cris tests new models
(8:51) who uses Runway?
(11:07) how Runway teaches new users
(14:38) does UI matter?
(29:39) what are the next frontiers for video models?
(33:02) allocating resources for the research time
(39:45) how Cris thinks about pricing
(42:40) one video model to rule them all?
(44:54) incorporating IP
(53:15) over-hyped/under-hyped
(54:00) biggest surprises while building Runway
(55:42) advice for art students
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
This week on Unsupervised Learning we had Krish Ramineni, Co-founder & CEO of Fireflies on to talk about how he sees AI changing the way we work and conduct meetings, and share his biggest learnings around AI in building Fireflies.
(0:00) intro
(1:28) how will AI change meetings going forward
(4:30) Fireflies’ capabilities
(8:45) how do new models impact Fireflies?
(11:46) shortcomings of current models
(17:04) Krish’s lack of belief in fine-tuning
(26:46) dealing with high inference costs
(34:27) taking on incumbents
(41:17) what metrics matter most to Fireflies
(47:03) how is Fireflies so fast?
(54:53) over-hyped/under-hyped
(55:36) biggest surprises in building Fireflies
(1:01:42) Jacob and Rashad debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
(0:00) intro
(2:27) Adobe’s new AI tools
(4:40) best uses of Adobe’s AI features
(7:22) educating users
(9:28) will the future have one model or thousands?
(11:01) Adobe building their own models
(15:12) goals for video generation
(19:40) hyper-personalized media
(22:11) AI music
(26:30) pricing for AI features
(28:02) biggest surprises in building AI features for Adobe
(30:58) most exciting AI startups (KoBold AI)
(32:13) where else would Scott build in the AI world
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 intro
0:49 Mikey’s favorite Suno songs
5:35 who uses Suno?
8:08 teaching people how to use Suno
10:15 new ways to prompt models
13:42 the social aspect of Suno
18:03 how does Suno approach pricing?
19:55 model eval
23:45 how models can improve
25:16 how is Suno so fast?
27:08 handling usage spikes
34:46 raising $125 million
41:18 IP partnerships
46:13 over-hyped/under-hyped
47:14 biggest surprises in building Suno
51:13 generating an Unsupervised Learning theme song
53:44 Jacob and Rashad debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
Eric's new podcast and newsletter:
ericriesshow.com
ericries.carrd.co
0:00 intro
0:33 does The Lean Startup apply to AI
4:34 thinking about defensibility
9:10 best way to get caught up on AI
11:34 starting Answer.ai
23:48 efficient fine-tuning of Llama 3
38:21 Jeremy's issue with the new AI legislation
48:27 over-hyped/under-hyped
48:53 most exciting AI startups
55:37 Jacob and Jordan debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 intro
1:02 advice for young coders
3:49 AI products at Sourcegraph
6:32 the current state of AI coding
12:48 what happens when a new GPT model comes out?
20:31 what types of developers benefit from these AI tools?
31:00 how important is inference cost?
35:55 how does Sourcegraph structure AI teams?
41:51 what metrics does Sourcegraph use to evaluate their products?
50:26 making RAG better
57:19 getting ahead of the agentic future
1:05:29 will there be more or less engineers in the future?
1:14:14 over-hyped/under-hyped
1:17:20 surprises during the Sourcegraph journey
1:18:50 cognition buzz and Devin
1:27:12 Jacob and Pat debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint
0:00 intro
0:35 what was it like when ChatGPT came out?
6:34 Edo’s favorite applications built on Pinecone
10:39 will we see more image and video applications in 2024?
15:06 best ways to deal with hallucinations
18:12 the evolving landscape of vector databases
20:36 if Edo had to build a product, what would his stack look like?
31:54 helping clients versus letting them figure things out
36:47 moving to a serverless model
40:42 what areas of AI should new startups target?
45:28 Amazon Sagemaker
50:41 over-hyped/under-hyped
51:28 biggest surprises while building Pinecone
56:14 Jacob and Pat debrief
With your co-hosts:
@jacobeffron
- Partner at Redpoint, Former PM Flatiron Health
@patrickachase
- Partner at Redpoint, Former ML Engineer LinkedIn
@ericabrescia
- Former COO Github, Founder Bitnami (acq’d by VMWare)
@jordan_segall
- Partner at Redpoint


