Pairing two AI agents to collaborate often fails. Because they share the same underlying model, they tend to agree excessively, reinforcing each other's bad ideas. This creates a feedback loop that fills their context windows with biased agreement, making them resistant to correction and prone to escalating extremism.

The key challenge in building a multi-context AI assistant isn't hitting a technical wall with LLMs. Instead, it's the immense risk associated with a single error. An AI turning off the wrong light is an inconvenience; locking the wrong door is a catastrophic failure that destroys user trust instantly.

Unlike humans who have an intuitive sense of when to stop searching, agents can get stuck in expensive, fruitless loops trying to find information that may not exist. Teaching models the judgment to abandon a task is a new and vital frontier for reliable agentic AI.

Even models with million-token context windows suffer from "context rot" when overloaded with information. Performance degrades as the model struggles to find the signal in the noise. Effective context engineering requires precision, packing the window with only the exact data needed.

To avoid context drift in long AI sessions, create temporary, task-based agents with specialized roles. Use these agents as checkpoints to review outputs from previous steps and make key decisions, ensuring higher-quality results and preventing error propagation.

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.

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.

A critical learning at LinkedIn was that pointing an AI at an entire company drive for context results in poor performance and hallucinations. The team had to manually curate "golden examples" and specific knowledge bases to train agents effectively, as the AI couldn't discern quality on its own.

There's a tension in agent design: should you prune failures from the message history? Pruning prevents a "poisoned" context where hallucinations persist, but keeping failures allows the agent to see the error and correct its approach. For tool call errors, the speaker prefers keeping them in.