YoutuBlog (HackerTube): YouTube-Vorträge und Podcasts lesen
Neueste Artikel
6.758 ArtikelSortieren
41:04ENHow Google tests Gemini 4 and uses agents to maintain codeTulsee Doshi, Smitha, Rody Davis, Luke, Mollie, Billy, Annie · Google DeepMindTulsee Doshi describes staged releases and long-running work; a demo audits web, Android and iOS code with subagents.
52:08ENWhen coding agents overwhelm a self-hosted model, reject earlyHarshul Jain, Parv Khatri · AudibleAdmission control can protect GPU capacity, but benchmark it against every offered request—including the ones rejected.
22:01ENZach Lloyd: make the software factory measurable and editableZach Lloyd · WarpWarp’s CEO proposes recording agent runs, scoring failures and testing changes to the workflow on real tasks.
17:27ENFix a tangled app by specifying its states before promptingDevin Chasanoff · GoogleDevin Chasanoff maps a workout screen’s state, transitions and tests before asking an agent to change it.
27:35ENUber traded an always-green trunk for faster AI-scale mergesPreetam Dwivedi, Manjari Akella · UberIts merge queue now relaxes some conflicts and validates the combined state after landing, with a verified-good fallback.
56:04ENHow Long-Term Investors Balance AI Ambitions With LiquidityGast: Rohit Sipahimalani, Jane Dietze, Shikha Gupta · Moderation: Drew McKnightTemasek, Brown and Australia’s Future Fund describe different approaches to AI exposure, private markets and preparing for shocks.
15:25ENHassan and Omar El Mghari: When an AI financial forecast is right for the wrong reasonsHassan El Mghari, Omar El Mghari · Together AIA Costco membership-fee model passed basic checks and landed close to reported results, but an audit uncovered offsetting assumptions.
27:58ENShould AI agents get your personal data? A debate over trustGast: Meredith Whittaker, Neil Chilson · Moderation: Jeff BermanSignal’s Meredith Whittaker warns that broad agent access can expose others; Neil Chilson argues for limited, verified use.
14:38ENJemima Abu: Give coding agents accessibility rules from the startJemima Abu · CAISA product engineer shows how semantic HTML, reusable components and explicit agent skills can help teams build more usable interfaces.
21:12ENTom Moor: Give coding agents the product context that code cannot showTom Moor · LinearLinear’s workflow carries customer requests, past decisions, review history and human feedback from intake through implementation and review.
15:35ENMislav Ivanda: Choose an execution boundary before an agent frameworkMislav Ivanda · DaytonaThe Daytona engineer compares local approvals, embedded runtimes and sandboxed machines for running agent-generated code.
2:07:25ENBugzy Malone on What Comes After SuccessGast: Bugzy Malone · Moderation: Chris WilliamsonThe Manchester rapper reflects on childhood, vulnerability, creative ambition and why building a life matters as much as making it.
21:47ENFable 5.2 won four creative tests; Opus resisted pressure betterPeter Yang compared five tasks and found Fable stronger at creative outputs, while Opus more often challenged misleading pushback.
15:21ENAshwin Paranjape: Testing financial agents against earnings consensusAshwin Paranjape · Samaya AISamaya compares seven models on more than 450 earnings releases using time-limited data, a bias-corrected baseline and several scoring methods.
17:57ENMaty Bohacek: Can AI agents impersonate people to pass verification?Maty Bohacek · Stanford UniversityA controlled vault experiment tests whether agents use available audio and video impersonation tools, and how safety alignment changes the result.
50:17ENMistral’s Arthur Mensch on sovereign AI and the compute bottleneckGast: Arthur Mensch · Moderation: Harry StebbingsThe CEO says customers need model choice and continuity, while European infrastructure must scale ahead of demand.
18:56ENPratyusha Duvvuri: Making AI token spending understandable to engineering and financePratyusha Duvvuri · CiscoA Cisco engineering lead explains why provider bills, team ownership and return on investment require different views of the same AI usage.
1:12:41ENWhy a Fluent AI Answer Can Still Be WrongCynthia Bailey · Stanford UniversityCynthia Bailey introduces next-token prediction, training and context to explain both the usefulness and limits of chatbots.
45:21ENTheo: AI coding makes maintaining the development environment the jobTheo Browne · Ping LabsFrom Twitch to T3 Code, Theo argues that engineers must remove the friction and feedback failures that block agents.
16:57ENFable 5.2 excels at ambitious builds, but its token costs add upEvery’s early tests found strong coding and creative work; a single documentary run used 1.4 billion tokens.
19:24ENRichard Feldman: keep the agent conversation with the codeRichard Feldman · DeltaA shared thread can preserve decisions, reviews and handoff context that otherwise gets lost between teammates.
1:09:35ENAn AI Agent Is a Loop, Not Just a Language ModelDiyi Yang, Michael Ryan, John Yang · Stanford UniversityDiyi Yang explains how reasoning, memory, tools and an environment fit together—and why task success alone is not enough to evaluate an agent.
15:28ENRyan Carson: One manager agent coordinates 20 coding agentsRyan Carson · UntangleUntangle delegates coding, review and merge coordination to agents, but its founder watches each feature before approving shipment.
1:50:34ENA financial AI agent must show its workFrancesco Galletta, Felippe Felisola Caso · MonzoA wedding-budget demo pairs tested calculations with evidence, decision gates and independent checks.
Mehr laden24 von 6.758 angezeigt









