Codex Workshop: Memory, Skills and Long-Running Workflows

AI Engineer1h 15m
Jason Liu against a black background beside the blue-and-white headline “LONG-RUNNING CODEX”.
0 comments · 0 votesOpen discussionClose discussion
Sign in to join the discussion

    Jason Liu presents his personal approach to using Codex as a general work assistant. The workshop connects voice input, project notes, reusable skillsAn AI agent skill is a reusable package of instructions, resources, and tool guidance for performing a bounded kind of work. and computer-use tools with practical examples of research, triage and artifact creation. His demonstrations and productivity reports describe his own setup rather than guaranteed results.

    The later sections organize work into persistent threads and monitoring loops, with explicit goals, progress checks and stopping conditionsAgent completion verification checks evidence that an agent achieved the requested outcome rather than merely producing intermediate work.. Jason Liu explains how project-level instructions and a shared repository can preserve contextContext engineering designs how relevant instructions, evidence, state and tool information are assembled for a model's task., while acknowledging that some experimental bookkeeping is costly and may not be useful to everyone.

    The question-and-answer sections address permissions, security boundaries, memory behaviorAI agent memory is stored information that an agent can retrieve and use across steps, sessions, or changing contexts. and reasoning effortReasoning effort is the amount of internal computational work an AI model applies before producing an answer or action.. Jason Liu cautions against assuming that the highest reasoning setting is always appropriate and shows imperfect outputs that still need review. Brief audience contributions are not treated as additional principal speakers.

    Original YouTube thumbnailWatch on YouTube

    Video topic coverage

    • Agents & workflows100
    • Coding & tools25
    • Models & learning25
    • Infrastructure & compute0
    • Research & evaluation25
    • Safety & security50
    • Business & productivity75
    • Creative & media0
    • Robotics & perception0
    • Society & policy0

    Share this video

    Jason Liu against a black background beside the blue-and-white headline “LONG-RUNNING CODEX”. With WWW.ARTIFICIAL-INTELLIGENCE.VIDEO, 24 July 2026 and duration 1h 15m.

    Jason Liu shows how memory, skills and persistent threads can support work beyond coding, while emphasizing clear goals, verification and proportionate reasoning effort.