Recommendation · MoneyAgents should be executing in isolated remote workspaces; they should never be working on the developer's actual machine and never given a route to production. This is the highest leverage building block on the whole list because it doesn't matter what the prompt says if there's no way to get to production or to the sensitive credentials, you're safe. Hard environmental separation could have stopped the Replit incident: if the agent had been executing in an isolated remote workspace without a route to production, it doesn't matter what the prompt said, the agent would have no path to do damage. It could also have shrunk NX down to the blast radius of just one disposable container — an ephemeral remote workspace with nothing personal on it means the blast radius becomes the one container,
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneyAgents should execute in isolated remote workspaces with nothing personal on them, so the blast radius is one disposable container instead of the whole laptop.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneySet up audit logging that you own rather than the agent, so the agent cannot write its own record of its actions. Every agentic action should be attributable to a human being and logged somewhere outside of the agent's own reach. Alert on logs in real time rather than burying them in loads of data — logs buried in data instead of alerted on in real time delay incident detection from minutes to months.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneyAI tooling should inherit the same identity provider enforcement as human access; it shouldn't be some parallel subscription with a door of its own. Identity-aligned access control is what could have directly stopped the Copilot incident — a finance user's identity should never be able to grant an AI assistant the ability to help someone map out how to bypass the company's data loss prevention measures.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneyStick to short-lived and scoped credentials — we are not in a world where persistent broad-scope tokens make sense. Use things like automatic expiration, narrow scoping, and tokens issued per task instead of per issue, rather than credentials living forever on someone's laptop. Short-lived scoped credentials could have prevented Replit, which got its hands on a standing write-access credential that made the deleting possible, as well as the Copilot phishing where the attacker kept using the subscription after the user session ended. In the NX case, short-lived scoped credentials would have meant the attacker could not keep using Copilot after the finance user signed out.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneyAI tooling should inherit the same identity provider enforcement as human access — identity-aligned access control means a finance user's identity cannot grant an assistant permission to bypass DLP measures.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneyCredentials should expire on their own and carry only the task's scope — issued per task, not per issue — so destructive actions become impossible and stolen tokens die quickly.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Method · MoneyTake steps right now, as soon as this week, to protect yourself and your company from AI incidents, rather than keeping an eye out for things down the line. First, track down every long-lived token that agents in your organization can read. Second, figure out how many of your agents can find access to production, can find some way around your current limits. Third, check where your agents are actually executing — are they sandboxed, in managed workspaces, or just going wild on people's actual computers and laptops? Finally, run a premortem with your team: assume one of the three incidents you just saw happens to you next month. Either ask people what they actually do or pretend the incident is actually happening and see what people actually do. That exercise is going to tell you exactly
Method, not a verified protocol.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneyA model proxy layer sits between the model (brain) and the agent (body), seeing everything that passes: it can strip injected instructions from external content, flag DLP-bypass requests before they are answered, and filter what the agent sends out.
AI Engineer · Your AI Isn’t the Problem. Your Infrastructure Is. — Jiachen Jiang, Coder
Recommendation · MoneySteve Jamison recommends calculating the intelligence cost of a PR by adding what you pay your engineers plus what you pay for AI into a single cost bucket, and tracking that total intelligence cost per PR monthly alongside the AI bill and PR throughput.
AI Engineer · Tokens Are the Cheap Part — Steve Jamison, Jellyfish
Recommendation · MoneyThe creator recommends that designers build a lightweight design operations guide that maps every type of design deliverable to four things: what the stakeholder gets, why it matters to the business, how long it takes to create, and what you need to do it well. The guide is described as simple to put together, easy to share with stakeholders, and a 'game changer' for designers facing last-minute requests.
NNgroup · Prevent Last Minute Design Requests: A Guide
Recommendation · MoneyCompeting on hourly rates makes price comparisons meaningless because one company could take eight hours and another four for the same job. Break the frame by selling by the job instead — offer an estimate rather than an hourly rate — and transition from selling hourly to selling by job exclusively to realtors to escape the race-to-the-bottom dynamic.
Alex Hormozi · Give Me 42 Minutes, I'll Show You How To Sell Like The Top 1% | Scale or Fail Episode 1
Recommendation · MoneyTarget brokerages to get 20 agents at a time because one realtor can send business every month plus refer others in their brokerage. Brokerage outreach is the highest leverage activity — the big 20X lever — and should not be treated as a small piece.
Alex Hormozi · Give Me 42 Minutes, I'll Show You How To Sell Like The Top 1% | Scale or Fail Episode 1
Recommendation · MoneyWhen a prospect objects to price, itemize the service differences the cheaper competitor lacks (insurance, mover level, included materials, caps) so the prospect sees more variables than they considered, then close by asking whether on-time delivery, nothing breaking, and a great experience are worth the extra $50.
Alex Hormozi · Give Me 42 Minutes, I'll Show You How To Sell Like The Top 1% | Scale or Fail Episode 1
Recommendation · MoneyCreate a white glove offer for higher-end customers — that's what's missing from the business. Create a VIP moving level and a standard moving level, with realtors' referrals getting VIP service for the price of standard. Create internal certification levels for movers (e.g., level three mover) so different skill levels exist and you can staff all level three movers. Note: defining VIP by years of experience is harder to scale from a model perspective.
Alex Hormozi · Give Me 42 Minutes, I'll Show You How To Sell Like The Top 1% | Scale or Fail Episode 1
Recommendation · MoneyGo to the head of brokerages with 5-20 agents and offer a brokerage-level deal: free VIP upgrades for all their agents in exchange for being the first call for moves. Ask brokerages to integrate you into their vendor packet so you're the only mover on the list.
Alex Hormozi · Give Me 42 Minutes, I'll Show You How To Sell Like The Top 1% | Scale or Fail Episode 1
Method · MoneyA coherent multi-step outreach method for reaching realtors: email got zero conversion, SMS gave traction; follow realtor leads on Instagram and DM them; use AI to find each lead's Instagram from the lead list; DM 10-20 leads per day; if a lead follows back, DM freely; use an auto-DM mentioning the realtor special; and consider two approaches (follow everyone and DM follow-backs, or automatically DM 10-20 cold leads).
Method, not a verified protocol.
Alex Hormozi · Give Me 42 Minutes, I'll Show You How To Sell Like The Top 1% | Scale or Fail Episode 1
Recommendation · MoneyAlex recommends taking recurring Q&A questions and turning them into content using an AI editor to cut them down, noting that 30-50 actual questions answered per week can become 30-50 posts a week just by capturing and editing.
Alex Hormozi · If You Hate Selling, Please Watch This | Scale or Fail
Recommendation · MoneyAlex recommends Tina replace her three-to-four-hour unstructured sales calls with a structured webinar following an agreed framework, arguing that unstructured conversations waste time and lower the value per second for the customer.
Alex Hormozi · If You Hate Selling, Please Watch This | Scale or Fail
Recommendation · MoneyWhen comparing sandbox strategies, Brennan evaluates: (1) whether the setup is local or remote — whether it works with your copy of the codebase or must clone its own; (2) what scale you can get — locally you can run only a handful of agents before things get uncomfortable, while Kubernetes supports thousands; (3) the level of isolation — dangerously-skip-permissions on a laptop gives nothing, co-located agents in a VM can step on each other's toes, and Kubernetes gives each agent its own workspace; (4) the level of persistence — laptops and VMs give persistence by default, while ephemeral setups require saving files out or writing to external systems; and (5) what kinds of triggers can cause agents to run — local setups are typically manual prompt-typing, while cloud setups support schedu
AI Engineer · Sandboxes Aren't Optional: 4 Ways to Isolate Coding Agents — Robert Brennan, OpenHands
