Eric Kim

@kevinolearytv As a slightly autistic engineer this is very concerning for me :')

The world has changed. If you know how to create, post, and understand platforms, you’re valuable. Period. The next generation isn’t just engineers. They’re storytellers, creators, and operators of attention.

@ollysmyth_ :')

@eric_nakji Minor?

storytelling is the hottest skill in tech right now. tech workers who can build a product are dime a dozen. but tech workers who have mastered the art of storytelling are extremely, extremely rate. now is the time to focus on these skills: - writing - presenting - social media

@Amplitude_HQ Clients who are building their own harness please reach out!

@Amplitude_HQ MCP now has progressive tool discovery mode! The default will stay with the raw list (since popular clients have dynamic tool discovery), but this should reduce token consumption by 110x 😱 lemme know if you want to try it out!

Add Amplitude analytics to your stack from the CLI without leaving your code.

@0xferruccio time to get them for the entire team 😏

pillaged an aeron chair from a dead startup, so now when I lay back I got sequoia backing me

we are shipping and shipping. check out some of the new tools we added!

@frankdotlee @Amplitude_HQ The flag tools have been incredibly helpful internally!

we’ve shipped a bunch of incredible new tools + skills using @Amplitude_HQ's mcp and i wanted to highlight a few killer use cases i’ve been running as a pm/engineer/builder! /replay-ux-audit + /debug-replay 👀🎥 we added two new session replay tools (list_session_replays and get_session_replay_events) which allows you to more effectively search across your session replays AND provides the agent access to the underlying events and rrweb data. this allows your agents to query session replays then analyze what’s happening in your replays. /monitor-ai-quality + /analyze-ai-topics 🤖📊 if you’re using our new agent analytics product, we’ve added mcp tools to query your LLM sessions, spans, conversations, and tagged traces. this enables your agents to easily browse your agent interactions, synthesize wha the top error categories are or main topics customers are talking to your agents about, then quickly iterate in your ADE/IDE on agent + prompt improvements. /review-agent-insights 💡 if you’re a heavy amplitude agent user, we’ve added a get_agent_results tool that lets you query all of your amplitude agent previous conversations and the artifacts attached to them. so if you have a bunch of dashboard and session replay monitors, you can stack workflows on top to triage and fix those issues. /diagnose-errors 🚨 fun fact: if you’re using autocapture, we track a bunch of rich web vital data like network requests, status codes, logged errors, and error clicks data as different amplitude event types. so you can automate workflows using an agent to auto-investigate web failures then fix those issues. get_flags, create_flags, update_flag 👨🏻‍🔬🧪 if you’re using amplitude for feature flagging, you can create, edit, and retrieve your feature flags all directly via MCP instead of having to manually configure in the UI anymore. we’ve been using this to scale out the number of experiments we’ve been running internally! give these a try and lmk if there’s other tools or skills we should be adding! 👀

shhhh @Amplitude_HQ now has Machine 2 Machine connection support via Service Accounts in beta! 👀 People wanted access to Ampli MCP using their own agents without interactive user consent. Lemme know if this would be useful for your own workflows!

@flowstated yall just playing with my emotions now 🔥 very cool! would be sick if there was some indication of estimated impact for changes

cursor now has design mode (⇧+⌘+D) - click to edit, drag to draw - shift + drag to box things in - add directly to chat with ⌥+click

@_bgiori I actually prompt my agents to add comments lol, I'd rather go back and delete but have it more readable to start

I noticed that Opus 4.6 went from adding comments on almost every line to never adding a single comment anywhere. I miss the comments to a certain extent...

@frankdotlee @Amplitude_HQ Seriously lol just bring it to where I'm already working

@eric_nakji @Amplitude_HQ i'm tired of clicking UI as well for this 😁

A small "update" lol, but you can now update @Amplitude_HQ experiments using the MCP! Agents can now monitor and update experiments, and then manage roll outs. Pretty nifty

workflows using a mix of different platforms and tools to help build! I literally start and end my day with a variety of these tools. Opportunity discovery and feature release management has become so much faster

Announcing Builder Skills A library of 50+ ready to use workflows you can run inside Claude for product, growth, analytics, and engineering

@thdxr hit or hit up 🤔

the hardest part about having a remote team is you can't hit your employees

@cdethiersant @Amplitude_HQ Honestly love this! Would love to know if there are any other workflows you would want to build and are blocked / having trouble building!

I had fun building a Slack bot to help maintain our @Amplitude_HQ event naming convention. It validates event names against our official object list, checks Proper Case, prefix casing etc... and even automatically creates an Amplitude branch in the background once approved

@geoffintech Not sure if I agree. If anything, because MCP tools are more workflow oriented and you can also look at tools called in a session being used flexibly without a set UI it can be more revealing. IMO sometimes the given restrictions in the UI hide what users truly want to do.

API usage through MCP/CLI has accelerated in the last months. This trend creates a paradox for PMs. Your product is getting more usage, but you understand less about how it's being used. API logs show you what was called, not what was built. That's a dangerous blind spot. The real threat isn't going headless — it's losing the feedback loop. Your power users are now building custom apps in Claude Code that you'll never see in your analytics dashboard. They're solving problems your product should have solved. The race isn't against Anthropic, it is against your own customers. They're already building the product they wish you had. Every day you don't talk to them, that gap widens. What does this mean: 1. Your MCP users are your most important research cohort right now — go talk to them 2. What they're building tells you what your product is missing. Build those capabilities natively, with the UX and reliability that a Claude Code hack can't match. This is your roadmap. 3. Ship it to the 90% of your users who will never touch an MCP and use your distribution to your advantage

Claude code source code has been leaked via a map file in their npm registry! Code: …a8527898604c1bbb12468b1581d95e.r2.dev/src.zip