Get your free personalized podcast brief
We scan new podcasts and send you the top 5 insights daily.
To manage context costs, Tasklet summarizes agent history with decreasing granularity over time. Recent interactions are sent verbatim, while older conversations have tool calls, thinking steps, and messages truncated or summarized. This is done in cache-aware buckets to minimize cost.
Related Insights
When an AI's context window is nearly full, don't rely on its automatic compaction feature. Instead, proactively instruct the AI to summarize the current project state into a "process notes" file, then clear the context and have it read the summary to avoid losing key details.
The new Codex app encourages a 'monothread' pattern where a single AI conversation is kept alive for weeks. Improved context compaction allows the thread's value to increase over time, moving beyond the old model of starting fresh for each task and creating a persistent, learning assistant.
Long, continuous AI chat threads degrade output quality as the context window fills up, making it harder for the model to recall early details. To maintain high-quality results, treat each discrete feature or task as a new chat, ensuring the agent has a clean, focused context for each job.
Despite massive context windows in new models, AI agents still suffer from a form of 'memory leak' where accuracy degrades and irrelevant information from past interactions bleeds into current tasks. Power users manually delete old conversations to maintain performance, suggesting the issue is a core architectural challenge, not just a matter of context size.
Don't pass the full, token-heavy output of every tool call back into an agent's message history. Instead, save the raw data to an external system (like a file system or agent state) and only provide the agent with a summary or pointer.
Long-running AI agent conversations degrade in quality as the context window fills. The best engineers combat this with "intentional compaction": they direct the agent to summarize its progress into a clean markdown file, then start a fresh session using that summary as the new, clean input. This is like rebooting the agent's short-term memory.
Long conversations degrade LLM performance as attention gets clogged with irrelevant details. An expert workflow is to stop, ask the model to summarize the key points of the discussion, and then start a fresh chat with that summary as the initial prompt. This keeps the context clean and the model on track.
The simple "tool calling in a loop" model for agents is deceptive. Without managing context, token-heavy tool calls quickly accumulate, leading to high costs ($1-2 per run), hitting context limits, and performance degradation known as "context rot."
Tasklet completely re-architected its agent, moving from feeding chat history into the LLM to treating the file system as the primary context. The agent now receives hints and pointers to relevant files, enabling it to handle infinitely long histories and larger contexts beyond the token window.
To make agents useful over long periods, Tasklet engineers an "illusion" of infinite memory. Instead of feeding a long chat history, they use advanced context engineering: LLM-based compaction, scoping context for sub-agents, and having the LLM manage its own state in a SQL database to recall relevant information efficiently.
