Free Daily Podcast Summary
Max Agency: Daily Summaries Delivered
by LangChain
Welcome to Max Agency, a podcast about how the best AI agents are actually being built. Hosted by Harrison Chase, CEO of LangChain, each episode goes deep with the builders designing, deploying, and learning from real agent systems in the wild. From architecture decisions to evals, tooling, and failure modes, Max Agency is for people who want to understand what it really takes to build useful agents.
Latest Episodes
The most recent episodes — sign up to get AI-powered summaries of each one.
- Aug 13, 20261h 8m
How Unify cut its AI agent costs 95% in two weeks
Connor Heggie spent his early career on a fifteen-person self-driving startup run like a research lab, then moved to Scale AI's mapping team, before becoming the co-founder and CTO at Unify. Unify builds agents for go-to-market teams, and for the last two years, one of AI's big mainstream narratives has been automating away the sales rep entirely. Connor and his team built the opposite: an agent that gives every sales rep "an engineer in their back pocket." He walks through how Unify's harness evolved from million-agent batch jobs to a chat product, and unpacks the engineering that makes it cost-effective to run at scale.–We also discuss:How Unify cut 90-95% of costs two weeks before launchThe 15-requests-per-second ceiling inside OpenAI's prompt cacheWhy your LLM judge must be a different model familyThe Speed Audit: why one at a time beats a table of 1,000Why Unify's subagents are just a function callWhat working on self-driving taught Connor about running evals–Timestamps:00:00 Introduction01:30 "Go-to-market is a search problem"06:10 The old workflow: drag-and-drop nodes over a million-row table09:40 Why the harness is similar to a coding agent's10:50 Running durable agents in the cloud without a full VM13:15 Giving models pandas-like superpowers over a live table15:50 Why Unify's subagents are just a function call18:25 The 15-requests-per-second limit hiding inside OpenAI's cache24:20 Optimizing for prompt caching hit rates28:20 Fork versus child subagents32:35 The Speed Audit: why one at a time beats a table of 1,00037:50 Locking memory to keys instead of letting the agent freestyle44:15 Why AI isn’t taking over sales49:00 What working on self-driving taught Connor about running evals52:46 Why your LLM judge must be a different model family54:45 Ditching full VMs for Monty, a Python REPL that suspends58:55 Semantic merge sort: why Connor is obsessed with RLMs59:53 How Unify cut 90-95% of costs two weeks before launch1:01:15 The case for "semantic linters" over skill files1:04:55 Why Unify runs mostly on OpenAI, their first investor1:06:12 Why 10x cheaper tokens still lose on tool efficiency1:07:24 Why open-source models don’t make economic sense (yet)–Referenced:AnthropicChatGPTClaudeClaude Fable 5Claude Sonnet 5Context HubDeep AgentsGLM-5.2GPT-5.4Helm.aiLangSmithOpenAIQuickJSRecursive Language Models (RLM)SalesforceScale AIScaling Managed AgentsUnify–Where to find Connor:Twitter/XLinkedIn–Where to find Harrison:Twitter/XLinkedIn–Where to find LangChain:Twitter/XLinkedIn–Send feedback or questions to maxagency@langchain.dev
- Jul 30, 202650 min
The misaligned incentives behind AI coding agents | Russell Kaplan, Cognition
Russell Kaplan is the President of Cognition, the $26 billion company behind Devin, the AI software engineer. He started his career in machine learning at Tesla's Autopilot team, then led the ML org at Scale AI, which had acquired his computer vision startup Helia. In this conversation, Russell unpacks why more and more engineering work is becoming "intelligence saturated," how the sidekick architecture inside Devin Fusion cuts cost without cutting quality, and why Cognition is underwriting a $10 million productivity guarantee.–We also discuss:Why mergeability is the next eval barWhen to use Fable vs GPT modelsHow smarter routing unlocked 35% better price performanceWhy letting users pick their own model is a UX bugWiring agents to be proactive instead of reactiveHow Cognition scores a session productive or unproductive–Timestamps: Introduction Launching Devin at 13% on SWE-bench When Devin became the number one committer to Devin Intelligence saturation: when speed and cost start to matter more When to use Fable vs GPT models Why mergeability is the next eval bar How open source maintainers shaped FrontierCode The telltale sign of agent-written code When per-person token spend starts to eclipse salaries Why letting users pick their own model is a UX bug How smarter routing unlocked 35% better price performance Why Cognition still trains its own models Deploying Cerebras at scale at 950 tokens per second The Tesla chip debate: does an inference chip need division? Wiring agents to be proactive instead of reactive Why "software factory" is the wrong term Why everyone is now a new grad Forward deployed engineering, and why ex-founders thrive The industry's incentive to get you tokenmaxxing How Cognition scores a session productive or unproductive Underwriting a $10 million productivity guarantee Using productivity dashboards to coach, not police–References:AnthropicCerebrasClaude FableClaude OpusCognitionCognition's AI Productivity GuaranteeDevinDevin DesktopDevin FusionDevin ReviewDevin Security SwarmEstimating the Productivity of an Autonomous AI Software EngineerFrontierCodeGPT-5.5GPT-5.5-CyberHarborLangSmithOpenAIOpenRouter Model FusionSWE-1.6SWE-benchTesla–Where to find Russell:LinkedInTwitter/X–Where to find Harrison:LinkedInTwitter/X–Where to find LangChain:WebsiteDocs–Send feedback or questions to maxagency@langchain.dev
- Jul 16, 20261h 17m
The best AI agents cost less than you think | Eno Reyes, Factory
Eno Reyes is the co-founder and CTO at Factory, a $1.5 billion company turning signal into deployed code inside some of the world's biggest engineering orgs. Before founding Factory in 2023, he was an engineer at Microsoft and then Hugging Face. In this conversation, Eno unpacks why the harness matters more than the model running underneath it, how to build your own 24/7 software factory, and why he's "bullish on humans in the loop for a very long time.”–We also discuss:Why product management isn't going awayThe engineer who lint-checks Factory's own agentsWhy coding agents might become the best general agentsThe platonic representation hypothesisWhy Factory might hide when "memory" is happeningBuilding Factory's universal meta harnessTokenomics, and how routing cuts the bill–Timestamps: Introduction What a 24/7 autonomous software factory actually means Why product management isn't going away Alvin, the engineer who lint-checks Factory's own agents "It makes me very bullish on humans in the loop for a very long time" The Disney Epcot analogy for rolling out AI Why coding agents might become the best general agents The case against a model-independent harness, and Eno's counter The platonic representation hypothesis explained Why model quirks are like being left or right-handed Why you could technically decompile Factory's entire harness Why memory might be AI's most overused word Inside AutoWiki and its Lore feature Agent readiness: the deterministic feedback agents need Missions: Factory's universal meta harness "It's kind of turtles all the way down": validating the validators Tokenomics: what missions cost, and how routing cuts the bill Why Eno is bullish on open models BenchBench, and why code review benchmarks might be broken–References:AiderAlvin SngAmpAndrej KarpathyAnthropicAnthropic: Emotion concepts and their function in a large language modelCursorDeep AgentsDeepWikiEpcotFactoryGLMHarborHugging FaceHuggingGPTKimiLangGraphLangSmithMiniMaxOpen Knowledge Format (OKF)OpenAIOpenRouter Model FusionRampSakana Fugu<
- Jul 2, 202650 min
The best AI agents are secretly teams | Ben Tannyhill, LangChain
Ben Tannyhill is a product manager at LangChain, where he's building LangSmith Engine—an agent that finds and fixes your agent's failures. Engine continuously analyzes your production traces, clusters them into actionable issues, and opens pull requests to fix them. Engine's architecture is a lot like an org chart: a main model delegating to a team of cheaper, faster sub-agents. It launched in public beta at Interrupt 2026, and in this conversation, Ben unpacks why it uses a sandbox as a tool, how the team turned it into a self-improving agent that learns from its own traces, and the hard problem of testing a fix before it ships.–We also discuss:Why Engine is "the agent for agent engineers"Making LangSmith agent-native with condensed trace viewsWhy the team keeps handing more control to the agentInside Engine's four sub-agents: the screener, verifier, and moreGiving Engine memory with an agent overview documentHow to keep an always-on agent from blowing the inference budgetWhere Insights, Polly, and Engine are converging–Timestamps: Introduction LangSmith 101 Why Engine is "the agent for agent engineers" Under the hood: Engine is a deep agent Clustering millions of traces with condensed views Why the team keeps handing more control to the agent Why Engine uses a sandbox as a tool Engine's four sub-agents and the org-chart analogy Evals for Engine: IssueBench, Harbor, and synthetic environments How Engine evolved: from noisy PRs to an issue inbox Inside Engine's memory: the agent overview document How to keep an always-on agent from blowing the inference budget What models Engine uses How Engine was rolled out: from Forge to public beta at Interrupt Inside the two teams building Engine Where Insights, Polly, and Engine are converging The missing piece: testing a fix before it ships Running a branched agent, and the write-access eval problem Using Engine as long-term memory Pointing Engine at coding-agent traces Running Engine on Engine: the meta self-improvement loop–References:AnthropicChat LangChainClaude CodeClaude HaikuClaude OpusCodexContext HubCredit GenieDeep AgentsGeminiGPT-5.5HarborHexInsightsInterruptLangGraphLangSmithLangSmith Chat (formerly Polly)LangSmith EngineLangSmith ObservabilityMintlifyOpenAIPalash ShahTerminal-BenchUnify–Where to find Ben:LinkedInTwitter/X–Where to find Harrison:LinkedInTwitter/X–Where to find LangChain:WebsiteDocs–Send feedback or questions to maxagency@langchain.dev
- Jun 18, 20261h 27m
The best AI agents are simpler than you think | Zack Reneau-Wedeen, Sierra
Zack Reneau-Wedeen is the Head of Product at Sierra, the conversational AI platform behind customer-facing agents for most of the Fortune 20. Before Sierra, he spent seven years at Google as the founding PM for Google Lens and Google Podcasts, then led product at Robinhood and CoinTracker. Sierra is mostly known for customer support, but Zack reveals how and why the company is building agents that span the entire customer lifecycle, from browsing and booking to sales and loyalty. In this conversation, he argues agentic commerce will be bigger than e-commerce, explains why he's a "monolith loyalist", and unpacks why, when a model looks dumb, the problem is usually you.–We also discuss:How Sierra's no-code layer compiles down to agent code, and back againWhy most multi-agent systems just ship your org chartInside Sierra's modular voice architecture: thinking, listening, and talking in parallelWhy Sierra built a PCI-certified stack for voice paymentsHow outcome-based pricing aligns incentivesWhy there's no breakout memory company–Timestamps:00:00 Introduction03:39 Analyze, build, release: how you build on Sierra07:54 Inside Ghostwriter11:04 Meeting models on their turf “80% of the time17:47 The one constraint Claude Code doesn't have19:35 Agent-to-agent: when an API call still beats MCP21:02 Why agentic commerce will be bigger than e-commerce27:31 Running models in parallel and ensembling transcription32:22 Inside the Agent Data Platform40:00 Context engineering: everything it needs, nothing more41:38 "Whenever you think the model's too dumb, the model's actually too smart"46:13 Why multi-agent systems are a trap48:44 Voice 101: latency, naturalism, and 60 languages56:11 When voice-to-voice passes 50%: the over/under57:03 Making memory a first-class primitive1:02:47 Why there's no breakout memory company1:08:02 Why the solution to all AI problems "is more AI"1:09:20 Why Sierra open-sources the tau-bench universe1:14:42 How outcome-based pricing aligns incentives1:20:26 Who thrives as a forward-deployed agent builder1:22:16 The Formula One analogy: why product is the bottleneck1:25:47 How Sierra interviews for agency–References:Agent2Agent (A2A) ProtocolAnthropicChatGPTClaudeClaude CodeClaude MythosClaude Opus 4.5CodexDeep AgentsGeminiHawaiian AirlinesLangGraphModel Context Protocol (MCP)Not Another Workflow BuilderRedfinSentryShopifySileroSiriusXMStripeTau-benchThinking Machines Lab–Where to find Zack:LinkedInTwitter/XSierra–Where to find Harrison:LinkedInTwitter/X–Where to find LangChain:WebsiteDocs–Send feedback or questions to maxagency@langchain.dev
- Jun 4, 202650 min
The tool design tricks behind Benchling's AI agents
Nick Larus-Stone is the Head of AI at Benchling, the R&D data platform that life science companies use to store and manage their experiments, samples, instruments, and analysis. Benchling has been around for since 2012. In October 2025, it launched Benchling AI, an intelligence layer with a chat interface, backed by an agent, that helps scientists find data, design experiments, and write reports. Nick came to Benchling through its acquisition of Sphinx Bio, the analysis startup he founded. In this conversation, Nick walks through what it takes to build agents for scientific work, and where the playbook from coding agents holds up and where it breaks down.–We also discuss:Why Benchling invests so heavily in getting clean data upfrontHow they cross-check answers between models to get more out of each oneWhy and how Benchling leans on production tracesWhere AI actually helps science today, and where it still gets stuckWhy understanding LLMs is closer to biology than software engineering–Timestamps:00:00 Intro01:22 What Benchling AI is, and the 14-year data platform underneath it04:36 Why a decade of structured data is a core advantage05:57 The architecture under the hood08:28 Similarities and differences compared to a coding harness11:14 Benchling’s multi-agent architectures14:36 Dealing with verifiable vs non-verifiable tasks16:19 Doing evals when clean benchmarks aren’t possible18:13 Context engineering: SQL vs. file-based harnesses22:11 Memory: agents that create and update their own skills25:30 What user education for scientists looks like30:33 Why understanding LLMs is closer to biology than software33:28 When will agents discover a novel cure for disease?44:58 The future of harnesses in science48:13 Why fine-tuning on biology hasn't beaten frontier models–References:Agent Skills (Claude Docs)Benchling’s Deep Research AgentClaude (Anthropic)Design of experiments (DOE)FDA Investigational New Drug (IND) applicationGemini (Google)Google AI co-scientistLangSmithModel Context Protocol (MCP)The Ralph (Wiggum) Loop (Geoffrey Huntley)Sphinx Bio–Where to find Nick:BenchlingLinkedInTwitter/X–Where to find Harrison:LinkedInTwitter/X–Where to find LangChain:WebsiteDocs–Send feedback or questions to maxagency@langchain.dev
- May 22, 20261h 14m
How Cogent builds AI agents that have to be right every single time | Geng Sng (Co-founder & CTO - Cogent)
Geng Sng is co-founder and CTO of Cogent, which builds autonomous agents that remediate vulnerabilities for enterprise security teams. Today, Cogent's agents process billions of security events per day, maintaining a live context graph of every asset and vulnerability across customer environments. In this conversation, Geng walks through Cogent's hot vs cold context split, the sub-agents that handle side quests, and the two graphs they run in parallel.–We also discuss:Why defensive security is harder for AI than offensiveUnder the hood of Cogent's three agentsInside Cogent's “read only” by-default sandboxesWhy graph databases don't scale for security dataCogent Research and the move into formal verificationWhy interactive agents need a deeper planning phase to one-shot–Referenced:Abnormal AIAmazon S3AnthropicBashChatGPTClaude CodeClaude MythosCodeMenderCodexCogentCursorGoogle DeepMindGPT-5.5-CyberJupyterLettaMozillaOpenAIOpus 4.6Opus 4.7Vercel–Where to find Geng:LinkedIn–Where to find Harrison:LinkedInTwitter/X–Where to find LangChain:WebsiteDocs–Send feedback or questions to maxagency@langchain.dev–Timestamps:00:00 Why mean time to exploit collapsed from years to minutes02:08 Inside Cogent's Agent Lake architecture05:11 Why Cogent rejected graph databases10:48 The trust ladder before agents touch production15:13 The three types of agents inside Cogent17:07 How Cogent sandboxes its agents19:16 Short-circuiting interactive agents with a deeper planning phase24:31 What to do when users believe agents too much31:21 Why sub-agents let agents go on side quests34:59 Two-tiered evals and the metric that catches bad prompts40:00 Cogent’s unique approach to context48:39 Cogent Research and the move into formal verification51:33 The single trait Cogent hires for54:00 Open-sourcing models within six months57:07 Why defensive security won’t be commoditized anytime soon1:00:51 The founding insi
- May 7, 202644 min
How Ramp built an AI agent that can think outside of tokens | Alex Shevchenko
Alexander Shevchenko is the head of applied research at Ramp, where he leads Ramp Labs – the team behind Ramp Sheets and a steady stream of public AI engineering experiments. Ramp Sheets started as an internal process mining tool that turned Loom videos of accountants into Markov diagrams, before evolving into the agentic spreadsheet editor that shipped in November. In this conversation, Alex walks through the architecture under the hood, why Ramp biases the agent toward Excel formulas over Python code gen, and two recent Labs experiments: Latent Briefing and a user-steerable revival of Golden Gate Claude.We also discuss:Under the hood of Ramp SheetsInspect, Ramp's internal coding agent, and the self-improving monitor loop it powersWhy finance professionals rejected code gen as too "black box"Why Anthropic models tend to excel at agentic spreadsheet manipulationThe case for putting the agent outside the sandbox, not inside itThe Loom-to-Markov-diagram process mining pipelineRLMs and how subagents can share memory in latent spaceLatent Briefing and KV-cache communication between subagentsReviving Golden Gate Claude with steering vectors on GemmaReferenced:Alex LevinsonAnthropicBen GeistClaudeEfficient Memory Sharing for Multi-Agent Systems via KV Cache Compaction (Ben Geist)GemmaGolden Gate ClaudeGraphvizInspectLatent BriefingLoomModalOpenAIOpusQwenRampRamp LabsRamp SheetsRecursive Language Models (Alex Zhang)RetoolSelf-maintaining Ramp SheetsSteer AIWhere to find Alex:LinkedInTwitter/XWebsiteWhere to find Harrison:LinkedInTwitter/XWhere to find LangChain:WebsiteDocsSend feedback or questions to maxagency@langchain.devTimestamps:00:00 Introduction01:13 The origin of Ramp Sheets02:27 The Loom-to-Markov-diagram process mining pipeline04:28 Why code gen approaches felt too "black box" to finance06:13 Meeting finance where they already are: inside the spreadsheet09:08 How far process mining got them10:31 Text descriptions and Graphviz DAGs as output12:41 Under the hood of Ramp Sheets14:52 Why the agent uses Python only as an escape hatch15:47 Why Anthropic models excel at agentic spreadsheet manipulation17:12 Frankensteining the OpenAI Agents SDK17:43 The Ramp Sheets UX and fast vs. expert mode19:58 Agent in a sandbox vs. agent with a sandbox21:55 Vibe evals with expert humans23:40 Inspect, the internal coding agent24:13 The self-monitoring loop and auto-PRs28:01 Other wacky experiments on Sheets28:4
Get Max Agency summaries in your inbox
Free AI-powered daily recaps. Key takeaways, quotes, and mentions — in a 5-minute read.
Get Free Summaries →Free forever for up to 3 podcasts. No credit card required.
You Might Also Like
Podcasts in similar categories.
All-In with Chamath, Jason, Sacks & Friedberg
Four tech investors discuss technology, markets, politics, and poker with candid, in-depth analysis.
Tomorrow, Today
Exploring how emerging technologies like AI, quantum, and space shape human potential and redefine the future of humanity.
Inside AsembleAI: DeepTech, AI & Science
Experts discuss AI's impact on business, product development, and leadership with researchers, founders, and executives.
Pivot
Sharp, unfiltered analysis of tech, business, and politics with bold predictions and lively debate.
Acquired
Explores the founding and growth stories of major companies and the strategies behind their success.
TBPN
Daily tech talk show featuring interviews with prominent tech leaders and condensed highlights of key discussions.
The Smoking Tire
Car enthusiasts discuss automotive news, racing, and projects with industry figures and guests.
Well There‘s Your Problem
A podcast examining engineering disasters and systemic failures through a leftist lens with humor.
About Max Agency
Welcome to Max Agency, a podcast about how the best AI agents are actually being built. Hosted by Harrison Chase, CEO of LangChain, each episode goes deep with the builders designing, deploying, and learning from real agent systems in the wild. From architecture decisions to evals, tooling, and failure modes, Max Agency is for people who want to understand what it really takes to build useful agents.
Customized Recaps
AI-powered recaps with compact key takeaways, quotes, and insights.
Straight to Your Inbox
Get key takeaways from Max Agency in a 5-minute read.
Save Hours Every Week
Stay current on your favorite podcasts without falling behind.
Frequently Asked Questions
What is Podzilla's Max Agency daily summary?
It's a free AI-powered email that summarizes new episodes of Max Agency as soon as they're published. You get the key takeaways, notable quotes, and links & mentions — all in a quick read.
How does the Max Agency podcast summary work?
When a new episode drops, our AI transcribes and analyzes it, then generates a personalized summary tailored to your interests and profession. It's delivered to your inbox every morning.
Is this an official Max Agency product?
No. Podzilla is an independent service that summarizes publicly available podcast content. We're not affiliated with or endorsed by LangChain.
Can I get summaries of other podcasts too?
Absolutely! The free plan covers up to 3 podcasts, any in the catalog. Basic raises that to 15, Pro to 40 and Max to 100. Browse our full catalog at /podcasts.
How often does Max Agency release new episodes?
Max Agency publishes biweekly. Our AI generates a summary within hours of each new episode.
What topics does Max Agency cover?
Max Agency covers topics including Technology. Our AI identifies the specific themes in each episode and highlights what matters most to you.
Start getting Max Agency summaries tomorrow morning.
Free forever for up to 3 podcasts. No credit card required.
Free forever for up to 3 podcasts. No credit card required.