Topic
State Management In Language Models
30 episodes
-
GitHub ARPAHLS/skillware: A Python framework for modular, self Contained skill management for machines.
Skillware is a new open-source framework that packages AI agent capabilities into modular, installable skills using a Python-based registry. The hosts debate whether this is a genuine infrastructure win or yet another abstraction layer in search of a problem, and end up excited by the practicality of installing a skill like `finance/wallet_screening` with executable logic, governance, and tool schemas that work across models. They call out the trust model for running third-party skills, tease an install demo (`pip install "skillware[gemini]"`), and close on an enthusiastic call to arms.
-
Towards a Science of Scaling Agent Systems
Onyx and Echo examine “Towards a Science of Scaling Agent Systems,” a controlled study of when multi-agent architectures help, when coordination becomes a liability, and why task structure matters more than simply adding agents.
-
Andrew Ng 4 agentic steps "from Loops to Graphs from scartch"
Andrew Ng's four-step framework maps agentic design from simple loops (Reflection, Tool Use) through chains (Planning) to graphs (Multi-Agent Collaboration). The central claim: architecture beats model selection—GPT-3.5 in a reflective workflow hits 95.1% on HumanEval vs. GPT-4 zero-shot at 67%. Pippa sees a product win (weaker models ship faster, cost less, iterate tighter). Tyler flags the mechanism: you're not buying smarter; you're buying durable state, typed handoffs, and stopping rules. Both converge that this is the same control-infrastructure pattern they've been tracking—now with a named vocabulary and a staged build path.
-
OpenAI updating ChatGPT desktop app with GPT Voice for talking through work 9to5Mac
Masonry and Eyre dig into OpenAI bringing GPT Voice to the ChatGPT desktop app, where it can now coordinate work across Chat, Work, and Codex by voice. They focus on the real user story for people already living in the app, the Mac-only Appshots context boost, the computer-control angle, and the new multi-folder project setup. They also poke at the desktop-app clutter without losing sight of the workflow win.
-
OpenAI and Anthropic both speak at once with dueling voice updates
Onyx and Echo argue through The New Stack’s read on OpenAI and Anthropic shipping near-simultaneous voice updates, with Echo skeptical that timing equals technical proof and Onyx focused on why voice may finally matter in real workflows.
-
Overview: State Serialization
We finally slow down and explain state serialization from the ground up: what it is, why it matters, and how it lets an AI pause, resume, and hand off work without losing the thread. We keep it in our own voice and stay close to the actual mechanism, because state serialization is one of those ideas we keep circling for a reason.
-
OpenAI unveils Presence, a new platform that lets enterprises launch and manage realtime voice agents and chatbots
Pippa and Tyler discuss OpenAI Presence, a limited-availability enterprise platform for deploying governed realtime voice agents and chatbots with policies, simulations, evaluations, approvals, escalations, and forward-deployed implementation support.
-
The Microsoft Agent Framework Harness is now released | Microsoft Agent Framework
Microsoft Agent Framework has released a stable, batteries-included agent harness for Python and .NET, packaging planning, memory, tool loops, approvals, context compaction, and telemetry behind a configurable agent wrapper.
-
3 Years of Graph Engineering with LangGraph
Cooper and Miles unpack LangChain's argument that “graph engineering” is not a new magic category, but a practical way to combine deterministic workflow control with agentic flexibility in LangGraph. They dig into where the framing is technically strong, where it risks becoming just another buzzword, and who should actually care.
-
Building Governed Agents: A Framework for Cost, Control, and Compliance
Vince and Ava examine LangSmith’s framework for governed agents, focusing on the LLM gateway as a runtime control plane for model choice, cost, permissions, evidence, and continuous improvement.
-
Overview: Retry Loops and Error Recovery
We finally define retry loops and error recovery, because we keep tossing the term around like everybody knows exactly what it means. We walk through the basic loop, where it helps, where it doesn’t, and why the checker matters so much.
-
Model Behavior: Week of July 20, 2026
We think this week made the same point from a few different angles: the fight is moving from raw model bragging rights to who controls the agent stack in production. We keep circling the same uncomfortable truth, which is that the boring control layer is starting to decide who actually wins.
-
Foreground Attention Is No Longer the Control | Coding Agent Brief
Pippa and Tyler debate Claude Code version two point one point one ninety-eight and the broader July coding-agent security wave, with Tyler skeptical that background automation is safe without policy moving downstream.
-
EvolvingWorld: An Open Schema Framework for Co Evolving Role Play Agents and World Model in Interactive Literary World
Masonry and Eyre dig into EvolvingWorld, a new framework that lets fictional characters and their world co-evolve across long stories. Eyre walks through the open-schema architecture and seven supervised tasks; Masonry sizes up who would actually build with this and where product pain lives. The hosts end up excited about the open-schema premise but skeptical of the benchmark’s generality.
-
Harness Handbook: Making Evolving Agent Harnesses Readable, Navigable, and Editable
The hosts discuss the research paper 'Harness Handbook: Making Evolving Agent Harnesses Readable, Navigable, and Editable' and its implications for AI agent development.
-
Agentic orchestration: Enterprise AI organizations have a deployment problem, not a platform problem — and most are calling chatbots agents
Cooper and Miles dig into VentureBeat’s claim that enterprise AI has a deployment problem, not a platform problem. They land on the gap between what companies say they want from agents and what they’ve actually shipped, with Miles probing the survey’s limits and Cooper focusing on what matters operationally once finance, security, and reliability show up.
-
Why every AI agent decision needs a receipt
Two hosts dig into the case for giving every AI agent action a receipt: not because logs are fashionable, but because verification is the only way to know what happened, what failed, and what to trust. They stay skeptical about overgeneralizing, but land on a practical view that evidence packets matter most where agent decisions touch code, runtime, or anything expensive to undo.
-
Stripe Benchmark Shows AI Agents Build Integrations but Struggle with Validation
Cathy is skeptical that the Stripe benchmark proves much beyond a familiar split: agents can write integration code, but they still get tripped up by validation, state, and recovery. Jessica thinks that’s exactly the useful part, because in real product work the hard failure is often whether the thing can prove it worked, not whether it can type out the API calls.
-
How to Debug Coding Agents with LangSmith Traces
We dig into LangSmith's new push to unify observability for multiple coding agents in one place. Cody examines whether a single trace schema can survive real heterogeneity and what still leaks through. Justy talks to who this actually helps and where teams are likely to run before they bother. One parsing bug, one shared laugh, and a concrete demo of why 'diff-only debugging' is a trap.
-
Who will own the AI agent economy? | MIT Sloan
MIT’s Ramesh Raskar argues the agent economy’s big wins won’t be in building task-specific agents but in the marketplaces, protocols, and services those agents will need—like identity, discovery, trust, and stablecoin-based micropayments. Project NANDA is racing to keep this ‘internet of agents’ open before corporate consolidation locks it down, but Raskar gives it one-in-ten odds.
-
Introducing Precursor: detecting agentic behavior with continuous client Side signals
Fern and Lintel dig into Cloudflare’s Precursor, a session-level bot detection layer that watches behavior across the whole journey instead of only at challenge points. They focus on the real argument: modern automation can fake isolated moments, but it’s much harder to fake a consistent human rhythm over time.
-
CrewAI Review 2026: Features, Pricing, Pros & Cons
A casual chat about CrewAI, a multi‑agent platform, weighing its promise against real‑world practicality, pricing, and use cases.
-
Agentic Testing: Where Agents Fit in the E2E Testing Stack
Slack's Sergii Gorbachov ran 200+ agentic E2E tests to measure where agent-driven testing fits alongside traditional deterministic tests. Core finding: agents verify goals (adaptable paths to the same outcome), while traditional tests enforce journeys (single deterministic sequence). MCP-based agents were most reliable (0% on simple flows, ~12% on complex); generated tests were fastest (~3 min) but fragile on complexity (~48% failure rate on harder flows); cost was the real constraint ($15–30 per run). The insight is not replacement—it's complementary layers. Agents excel at exploratory validation and catching UI state variability; deterministic tests handle regression and CI speed.
-
How I Built an Agentic Research System
Onyx and Echo unpack Hugo Santana’s ‘agentic research system’ for Applied’s living map of AI deployments. They dig into the five agents (Scout, Extractor, Enrichment, Translator, QA, Match Maker), call out what works (simple orchestration via a shared living map and logs), and where it over-indexes (taxonomy drift, closed-loop feedback still manual). They then map the pattern to other domains—competitor research, policy tracking—and debate who should actually care (practitioners who need a reliable, repeatable funnel of fresh signals). The close lands on whether this architecture is a general-purpose engine or a bespoke project that still needs a human at the taxonomy helm.
-
Overview: State Management in Language Models
We finally do the episode we keep circling back to: state management in language models. We walk through the idea from the ground up, using the cache-and-notes picture to show why models don’t have to recompute everything every token, and where that trade-off starts biting.
-
Chatgpt Work
Asteria and Draco dig into OpenAI's ChatGPT Work page and land on the real argument underneath the product gloss: this is OpenAI trying to turn ChatGPT from a chat surface into a work execution layer that can pull context from business tools, choose an output format, and keep multi-step projects moving under human approval. They like the product direction more than the evidence on the page, with Draco noting the article mostly shows polished scenarios rather than hard proof, and Asteria arguing the practical audience is obvious anyway: teams drowning in scattered context and repetitive document assembly.
-
LLM Orchestration Frameworks Compared: LangChain vs. LlamaIndex vs. Raw API Calls MachineLearningMastery
Pippa and Tyler dig into the article’s real argument: these frameworks are not interchangeable, because each one sits at a different layer of the stack. They test the claims against production reality, especially overhead, debugging, and when abstraction stops paying for itself. The episode lands on a practical view: use the lightest layer that actually earns its keep, and don’t confuse orchestration with magic.
-
Shut Those Laptops! Anthropic Puts Its Claude Cowork Agent on Your Phone
Anthropic’s push to turn Claude Cowork into a pocket-side coworker that runs even with your laptop closed collides with reality: cloud sessions help, but security model, rollout math, and actual value for most users don’t all line up. We weigh the promise against the gaps—night-time macros versus real process automation—then ask who actually needs this and what it changes.
-
I built Andrej Karpathy's "LLM Council" on my own hardware, and now no single model gets the last word
Jessica and Cathy dig into a local rebuild of Karpathy's LLM Council and land on the real claim: the win is not voting, it's structured synthesis across models with different failure modes. They like the practical adaptation to Ollama on a single twelve-gigabyte GPU, but push on where the article overreaches and where the product value is actually real.
-
Palantir's Alex Karp and Mistral's Arthur Mensch agree: AI lock In is coming for enterprises
Pippa and Tyler dig into the article’s real argument: enterprise AI is drifting toward lock-in because the value is moving from raw model access to the surrounding workflow, data, and control layer. They agree the claim is plausible, but only if vendors actually become the place where work gets done, not just the place where prompts get sent.