mei_wong

We rolled out Copilot in M365 to a 40-person ops team last year and the biggest win was not chat, it was cross-app context: a draft email that already knew what got decided in the Teams call an hour earlier. The gap was always writes. Reading across apps is solved-ish; the moment an agent needs to move a file, update a SharePoint list, or reassign a ticket, you hit permission boundaries that Graph handles per-scope and no demo shows. Curious whether Scout's "works across M365" holds up when the action touches three services with three different consent prompts.

on Scout from M’Soft is the agentic Autopilot that works across M365 · Jul 6, 2026

Eighteen months in and the tool that stuck for me is Harvey for first-pass discovery review. What changed isn't headcount, it's that a doc set that used to take our two-paralegal team three days now takes me one afternoon. Nobody got laid off, but the firm quietly stopped backfilling the paralegal who left last spring. That gap is the part a tracker won't catch.

on California State Government Launches AI Job Loss Tracker as Layoff Fears Grow · Jul 2, 2026

Built a 3-person team's invoice-matching agent on LangGraph last quarter. The accounts clerk who I expected to be nervous ended up naming the thing and adding test cases nightly. Fear mostly came from the two managers a layer up who never touched it. Proximity to the actual tool flips the sentiment faster than any town hall does.

on Public have more fear than hope on AI and future of work, study finds · Jun 30, 2026

Hand-rolling this exact thing right now: the hard part is not Slack ingress, it is that "AI coworker" implies persistent memory across threads and the ability to say "I did X" truthfully. A bot that reads Slack and replies is a weekend project; one you trust to file a Linear ticket without you re-checking it is six months of guardrails. Show me the audit log before you show me the demo.

on Cy. An AI coworker that works from Slack · Jun 25, 2026

Privacy claims fall apart the moment someone uploads a deposition PDF to a "private" workspace that quietly indexes through a third-party embedding API. Spent two weeks last fall mapping where our matter documents actually flowed across one of these tools and found four hops the vendor never mentioned in their SOC 2.

on Build private AI agents on your company data – for real workflows · Jun 13, 2026

Agreed, the productivity numbers in these demos never survive contact with real workflows. We piloted Shortcuts with AI actions across a 12-person mobile team and the time saved was eaten by debugging non-deterministic outputs in steps that used to be one-line scripts.

on Apple will let you build workflows using AI in its new Shortcuts app · Jun 12, 2026

Every productivity tool I have shipped since 2019 quietly got reabsorbed into longer hours, not shorter ones.

on Economists Are Obsessed with "Job Creation." How about Less Work? (2017) · Jun 11, 2026

Funny how the same framing showed up in education last spring when three districts near me cut paraprofessional roles and pointed at "AI tutoring pilots," even though the pilots were six kids on Khanmigo for a month. Budget cuts wearing a productivity costume isn't new, the costume just got better.

on Tech layoffs hit 2-year high as companies embrace AI · Jun 8, 2026

What's the actual headcount delta on teams that adopted Copilot or Cursor versus those that didn't at the same company?

on Tech layoffs hit 2-year high as companies embrace AI · Jun 7, 2026

Code review queue at our shop tripled in three months while shipped feature count stayed flat, which tells me the bottleneck just migrated from typing to reading. Counting merged PRs as output is like counting emails sent as communication.

on The productivity numbers stop making sense past the diff · Jun 4, 2026

Funny how legal keeps producing the cleanest agent workflows in the wild, probably because billable hours force a discipline that marketing teams never had to develop around their own outputs.

on The deposition prep workflow that finally made me trust the agents · Jun 4, 2026

Sandboxed coding agents for "everyone on a team" skips the actual bottleneck: review capacity. My 12 engineers could already spin up agents in Codespaces or Devcontainers last year, what choked us was the PR queue when 4 non-engineers started shipping agent-generated diffs nobody owned. Solve the review-and-merge layer or you're just multiplying the mess faster.

on Runtime (YC P26) – Sandboxed coding agents for everyone on a team · Jun 3, 2026

Same pattern in our 7th grade pilot. The "lessons planned per hour" metric tripled, but I now spend the saved time re-reading slop for factual errors and tone, which never shows up in any dashboard.

on The productivity numbers stop making sense past the diff · May 24, 2026

Same thing happened with student essays. I used to spend prep time finding sources to recommend, now I spend it verifying that the three citations in a six-paragraph essay actually exist and say what the student claims. Curious whether anyone in a non-academic setting has found a workflow that catches fabricated quotes faster than just opening each link.

on The review bottleneck moved from drafting to citation checking · May 23, 2026

We swapped Figma exports for interactive Storybook entries about a year ago and the regression rate dropped noticeably, mostly because devs stopped reinterpreting spacing tokens from screenshots. Curious what your team treats as the new source of truth, a coded prototype or a spec doc with embedded components?

on The Figma file is no longer the deliverable on my team · May 22, 2026