35,595 results · Most relevant
Rick Rubin describes the seed collecting phase as wide open with no deadlines, and says he is always in phase one — collecting seeds all the time, whether or not he is working on a project, noting anything that interests him or has potential.
Breaking away from the current topic to voice real-time observations about her is rated S-tier; observations can also become labels or hooks for later techniques.
The coach advises against sitting down right after opening a conversation. He recommends carrying the conversation standing up for at least the first minute (or one to two minutes) before pivoting to sit down, and only sitting once there's enough interest from the woman.
If the sales process takes too many calls, too long, or generates too many questions, prospects need more education ahead of time. Educating prospects before asking them to buy produces more money, more convicted buyers, fewer and shorter conversations, faster purchases, and more prepayment. The one downside is losing some people who don't make it through the education, but in in-person selling environments where prospects are already at the office, there is no downside at all.
Solve the supply constraint first, and then the other things on the pipeline can increase. More marketing on a supply constraint does nothing; it does not increase throughput. Only pursue additional marketing if you have bandwidth after solving constraints. Marketing only works if you have a way to capitalize on the demand it creates, though the more people know about your business, the more money you make.
The first sandboxing strategy people experiment with is running agents inside Docker containers on their laptop — this is how OpenHands originally launched, spinning up a new container per conversation. Containers are ephemeral, so if the agent runs rm -rf / you just kill the container and start again. You can selectively mount things from your local file system (like a Git repo), mount ports so you can view the app in your browser, mount specific secrets and environment variables, and run MCP servers locally or inside the container. It's nice because the agent works with your local copy of the code and you decide when to git push.
The next strategy is gluing together Claude Code and GitHub Actions: the agent runs in an ephemeral environment inside GitHub Actions, clones the repo, does its work, and everything goes away when the action spins down. You can run as many agents as you're willing to pay for, GitHub Actions has built-in support for things like a secrets registry giving nice isolated environments, and you can run agents on a schedule (e.g., a Monday 9 a.m. dependency update) or on events (e.g., agent code review any time a PR opens).
The principle of least privilege holds that both agents and humans should only have access to the things they need to do their job. An agent shouldn't have access to every secret or every file on your system; it should only have access to the GitHub repo it's editing and maybe a couple of secrets it needs to access a local database. When sandboxing, look at exactly what the agent needs to do its job, then create a sandbox with just those things — this limits the blast radius of a prompt injection attack or the agent getting confused.
Hormozi's third principle: leverage early success into bigger, later success. Successes are stepping stones you can ladder on top of. He illustrates with his own career: getting in shape early led to personal training clients, which became a gym, then multiple gyms, then Gym Launch, whose sale funded acquisition.com — every part of his career built by taking a stepping stone from the previous enterprise. He adds that the steps of the career ladder are set; the only variable is how soon you start climbing.
Hormozi scripts a conference talk approach: frame it as 'this is what makes shitty events vs great events,' build a slide deck of about 20 mistakes people make when buying event planning with proof or a testimonial under each mistake, then end with a soft CTA — 'hey, if you want the slides, go grab them' — via a QR code on screen. The opt-in form asks 'Do you run events? How many per year?' to qualify leads and build a list without pitching.
Once an agent is up and running in the cloud, all you have to do is start a session with the agent, send events, and get back events. You can steer a running agent by sending a new user.message event via send event, which gets queued up; you can send 10 messages at a time and it's fine.
If your company data is accessible via CLI with an API or via an MCP, you would just create an MCP, give it the authentication vault, and the agent would be able to call your MCP directly. If you don't want to expose your MCP to the internet, use MCP Tunnels, a feature that lets you create a tunnel from your network into Anthropic's so your MCP is not exposed to the internet. If MCP Tunnels is too big a barrier, use self-hosted sandboxes: bring your own sandbox peered with your network that has access to your applications and services, and the agent calls the sandbox you manage and host to access your resources.
Design agents to run on deterministic data with minimal reasoning — pull hard data from contracts, billing, and accounting systems rather than asking the agent to reason. Automated workflow on top of deterministic data beats heavy agent reasoning — the agent automates the workflow, it doesn't reason through it. When agents must reason (e.g., building pitch decks), bound the amount of reasoning and add guardrails to limit error surface.
For every LLM step, Flywire restricts four things: the network the agent can reach, the tools it can call, the skills it can load, and the MCP servers it can use — down to the specific tools allowed from each server. Standardized guardrails let engineers across SRE, security, IT, and support build custom workflow steps easily and safely.
Flywire enforces strict JSON schemas on every LLM step in Orbit workflows so that execution branches, error handling, and state transitions are driven by deterministic control flow rather than unpredictable string parsing; failed validations trigger a targeted retry or fallback pass automatically.
Flywire's evaluation harness isolates individual LLM steps instead of evaluating entire workflows end-to-end, injecting exact historical state and context into specific workflow nodes. This lets them evaluate model output quality, measure prompt adjustments, and catch performance regressions quickly and cost-effectively.
Flywire maintains strict isolation between engineering SDLC data, corporate operational data, and sensitive payment production data, and notes that managing data separation remains an ongoing responsibility when running agentic workflows.
Matt Ragone stresses that the agent never sends anything to a client; a planner reviews every single recommendation before a client sees it. In a regulated business giving advice to real clients, the review step is not optional.
Anthropic explained that when a managed agent session is waiting for a custom tool result, execution pauses and you cannot send messages to the session; sends are blocked for ordering purposes and attempting one returns a 400 error. If the model is doing other work (not blocked on a tool result), you can queue up messages.
To understand checkpoints and status for a long-running agent, use the session event stream via the events API under the session (also surfaced in the session tab).
Choose sources → to get a library of your own and a daily brief.