Résultats
28 articlesTrier
0:22EN
Theo - t3․gg· 12 sept. 2026Theo’s Take on GPT-6 Astra: A Next-Generation Model With Rough EdgesTheoTheo praises the model across coding and other tasks while warning that it can still over-engineer and get stuck in loops.
1:16:51ENTheo - t3․gg· 11 sept. 2026Fable 5.1 vs GPT-6 Astra: Theo on Coding, Computer Use and CostTheoTheo compares the models through coding projects, 3D demos and agent workflows, and explains why he uses each for different jobs.
49:22ENTheo - t3․gg· 9 sept. 2026Theo’s AI Agent Workflow: Parallel Investigations, Review Loops and Safer MergesTheoA walkthrough of how Theo uses agents, remote previews and release safety nets to keep coding while running a company.
28:36ENTheo - t3․gg· 7 sept. 2026Theo on Partial Understanding: Working in Large Codebases With AITheoReading Sean Goedecke on theory-building, Theo explains why engineers need useful mental models without memorizing an entire system.
44:12ENTheo - t3․gg· 4 sept. 2026GPT-6 Astra: Theo’s Early Tests of Computer Use, 3D and CodingTheoTheo reviews launch-era pricing and benchmarks, shows projects built with the model, and recounts a frustrating PR-review failure.
55:06ENTheo - t3․gg· 3 sept. 2026Fable 5.1 in Practice: Theo on Costs, Agent Workflows and PR QualityTheoTheo reviews the launch, prompting guidance and his early coding experiments, including an observational comparison of pull requests.
0:24ENTheo - t3․gg· 3 sept. 2026Theo’s Low-Cost Experiment Auditing T3 Code Pull RequestsTheoTheo describes using an inexpensive model to review recently updated pull requests and proposes a recurring readiness check.
38:05ENTheo - t3․gg· 3 juil. 2026Why Theo Moved His Agent Work to Linux MachinesTheoLong-running coding agents, laptop CPU load and slow small-file operations led Theo to build a networked fleet. He demonstrates SSH, a hardware KVM and T3 Code for remote work.
28:37ENTheo - t3․gg· 1 juil. 2026Theo Tests Sonnet 5 and Explains Fable’s ReturnTheoTheo compares model benchmarks and his own coding tests, finding promise in Sonnet’s subagent orchestration but questioning its task costs.
2:19ENTheo - t3․gg· 1 juil. 2026Theo on Sonnet 5: Agent Orchestration and Token CostsTheoTheo welcomes Sonnet 5's ability to coordinate subagents while questioning its token efficiency. He also discusses the introductory pricing announced at release.
19:24ENTheo - t3․gg· 26 juin 2026Theo on Google’s Coding Agents, Departures and Internal ExperimentsTheoTheo distinguishes model knowledge from agent behavior and argues that organizations need room for useful prototypes.
44:34ENTheo - t3․gg· 22 juin 2026Theo’s Six Ideas for Better Developer Tools and PlatformsTheoFrom safer package distribution to an open mobile platform and unusual AI benchmarks, Theo shares projects he hopes others will build.
0:26ENTheo - t3․gg· 22 juin 2026Theo on Agent Workflows: Don't Predefine Every PersonaTheoTheo explains his initial skepticism about orchestrating multiple coding agents and why he sees predefined personas as a poor fit for AI's dynamic nature.
23:21ENTheo - t3․gg· 20 juin 2026Theo: Think Bigger When AI Makes Building CheaperTheoTheo compares AI-assisted development with the cloud transition and explains why he is experimenting with a broader developer platform.
0:59ENTheo - t3․gg· 19 juin 2026Theo: Design Agent Feedback Loops Instead of Every PromptTheoAfter mixed results with early loops, Theo describes why agent-driven review and iteration changed his mind about prompting.
0:36ENTheo - t3․gg· 18 juin 2026Theo: Use One-Off Code to Coordinate Agent LoopsTheoA short example of an agent writing temporary code to trigger a workflow makes Theo rethink how models can prompt themselves.
24:44ENTheo - t3․gg· 18 juin 2026Theo on Coding Agent Loops: Review, Coordination, and CostTheoTheo describes moving from manually passing review feedback to dynamic agent workflows, including an overnight multi-PR experiment and its limitations.
2:48ENTheo - t3․gg· 17 juin 2026Theo on Claude Code Features He Wants Other Tools to AdoptTheoDespite his criticisms, Theo highlights skill scripts, context imports, side questions, workflows and remote control as ideas worth borrowing.
27:47ENTheo - t3․gg· 17 juin 2026Theo on Claude Code: Skills, Workflows, and Session ControlsTheoTheo walks through features he hopes other agent tools will adopt, including dynamic skill context and a code-generated PR audit.
2:30ENTheo - t3․gg· 12 juin 2026Theo’s First Look at Fable 5 and Its BenchmarksTheoTheo discusses the relationship between Fable and Mythos, benchmark refusals and early coding results.
0:38ENTheo - t3․gg· 12 juin 2026Theo’s Fable Workflow Hit Subscription Limits and Usage CostsTheoTheo describes a costly usage-based run and an unfinished workflow that exhausted two session limits.
30:40ENTheo - t3․gg· 12 juin 2026Theo on Agent Workflows, Usage Limits, and Building MoreTheoTheo demonstrates multi-agent PR reviews, ranked work queues and remote development while reflecting on cost, verification and sleep.
32:53ENTheo - t3․gg· 11 juin 2026Theo’s Fable 5 Review: Benchmarks, Real Projects, and LimitsTheoTheo tests a newly released model on coding and creative work, highlighting impressive results alongside failures, safeguards and cost.
20:29ENTheo - t3․gg· 3 juin 2026How Prompts Become Technical Debt in AI CodingTheoTheo examines outdated agent instructions, model-specific behavior and the value of regularly auditing project guidance.Charger plus24 sur 28 affichés