The author shares a year of experience using AI for data work, arguing that copying someone else's workflow is pointless — what matters is learning specific techniques. He's settled on DuckDB CLI + Claude as his stack because it reduces hallucinations and keeps agents focused on actual tools instead of generating buggy code.

• AI workflows are creative, not formulaic — tips and tricks transfer better than full process replication, similar to how watching a music producer's exact steps won't let you recreate their song
• DuckDB CLI commands paired with AI agents dramatically cut hallucinations because agents understand the tool's actual capabilities rather than inventing Python code that doesn't work
• Working in a modern terminal (Ghostty) with multiple windows beats IDEs for analytics work — it's faster, gives you exactly the tools you need, and agents are already built to work with CLI tools

Andrej Karpathy walks through practical daily AI workflows in a 2-hour video, covering model selection, reasoning models, code execution, and multi-chat memory — techniques most people never use. Someone extracted these methods into a Claude-specific guide with ready-to-use examples.

• Most people use only 10% of what AI models can do; this covers the remaining 90%
• Specific techniques shown include choosing the right model, deciding when reasoning models justify the cost, generating full research reports from single prompts, and automating code execution
• The guide translates Karpathy's video into Claude-specific features with immediately applicable examples

Flue is a TypeScript framework that turns large language models into autonomous agents by providing sessions, tools, sandboxed execution, and durable workflows. It adds subagents, observability, fault recovery, and Model Context Protocol integrations so you can run agents locally via CLI or deploy them on hosted runtimes.

• Flue turns simple TypeScript config (model, sandbox, skills, instructions) into a fully deployable HTTP-exposed agent, no separate infra work needed.
• Built-in durability tracks agent state across restarts so long-running workflows survive failures.
• Supports subagent delegation (e.g., one agent triages issues, another verifies fixes) for specialized multi-role workflows.
• Comes with a full ecosystem out of the box: CLI, runtime, SDK, Postgres/OpenTelemetry adapters, MCP tool auth, and integrations with Slack, Teams, Discord, GitHub, Braintrust, and Sentry.

This article lays out two guiding principles for PMMs: safeguard your unique skills (storytelling, judgment, strategic thinking, and stakeholder influence) and protect your cognitive abilities by using AI as a second step. It then offers a practical three-tier framework—execution, thinking, and scaling workflows—and advice on selecting high-impact use cases based on how you actually spend your time.

• Do the thinking yourself first, then use AI as a critic/amplifier—not the other way around—to avoid "AI brain rot" that erodes storytelling, strategic reasoning, and influencing skills
• Real insight still comes from live customer conversations and sales calls, not AI transcripts of them
• A three-tier framework works best: Level 1 (execution tasks like summaries/release notes) for low-risk time savings, Level 2 (thought partnership) for stress-testing messaging and strategy, Level 3 for high-stakes work still requiring full human oversight
• PMM's job is shifting from content producer to quality gatekeeper, strategist, and connector as AI floods the market with generated content

AWS Lambda now supports S3 Files, letting functions mount S3 buckets as file systems for direct read/write without object downloads. This simplifies stateful and AI/ML workflows by enabling shared workspaces and automatic checkpointing via durable functions.

• Lambda functions can now mount S3 buckets directly as file systems (S3 Files, built on EFS) instead of downloading/uploading objects in code
• Multiple Lambda functions can access the same S3 Files mount simultaneously for shared, real-time workspaces without custom locking
• Combined with durable functions for checkpointing, this avoids the 512 MB /tmp storage limit for stateful, multi-step workflows like AI/ML pipelines
• Setup works through the Lambda console, CLI, SDKs, CloudFormation, or SAM, with no extra fee beyond normal Lambda and S3 pricing