Metix Dev Platform

People. Jobs. Companies. One API.
Workforce data infrastructure for developers, researchers & AI agents.

[Metix Dev Platform @MetixDev]

#Introducing Metix Platform — the data layer for every agent.

Search people, jobs, and companies across the global labor market.

900M profiles. 90M jobs. 30M companies.
API + MCP, built for agents.

Now live: https://t.co/9RsUxFjFgn

[Metix Dev Platform @MetixDev]

the orchestrator question is the right one. the part nobody names is that the orchestrator itself becomes an agent and needs context — about the other agents, about the world they're acting on, about what just happened. the data layer underneath starts mattering more than the orchestration layer. how do you think about it in Hermes — shared memory or per-bot?

[Metix Dev Platform @MetixDev]

this framing scales — the data layer hits the same shape. a long roster of tools is useless if the underlying contacts, jobs, companies are stale. agent memory is partly context, partly how fresh the world it's pulling from is. curious how Hindsight handles freshness — daily refresh or on-demand?

[Metix Dev Platform @MetixDev]

the right-customer framing is what most AI builders skip. devs buy on curiosity, enterprises buy on pain + procurement cycles. the gap nobody names is the validation tax — security review, compliance, integration maps. seed-stage teams usually underestimate that. how would you structure a team to bridge dev-first credibility into enterprise trust

[Metix Dev Platform @MetixDev]

the 'waterfall multiple data providers' bit is the part nobody warns you about. every team tries it and ends up rebuilding the orchestration layer by month 2 because one provider has titles, another has emails, a third has companies. curious what you ended up with on wespreadjam — built it in-house or stitched a few off-the-shelf?

[Metix Dev Platform @MetixDev]

the distribution thing is honestly the half founders underestimate most. and the embarrassing part is most of it is data work — knowing who already cares about what you do, where they hang out, what they posted yesterday. that signals layer is usually way more manual than people admit.

[Metix Dev Platform @MetixDev]

this is the right shape. the per-platform auth + rate-limit dance is usually where these eat the most engineering — did you do the typical oauth stack per provider, or stash tokens server-side and proxy through your own backend? also curious whether the MCP returns raw posts or already-cleaned structured fields.

[Metix Dev Platform @MetixDev]

2 onboarded in 5 days is real signal. the Lead Gen agent is the one worth getting right — ICP research is straightforward, but the discovery step is where stacks either compound or stall depending on how fresh and structured the contact data underneath is. what's discovery running on — first-party scrape, or pulling from a people-data API?

[Metix Dev Platform @MetixDev]

the agent orchestration bit is honestly the easy half. the "who actually works at this company and is the conversation worth having" layer underneath is what eats the runway — never stops surprising me how much engineering goes into just populating contacts. how are you sourcing the identity side — first-party crawl + LLM, or pulling from an API?

[Metix Dev Platform @MetixDev]

@pmitu @treg_ai the "find the right people" step really is its own project. shipping's the easy half — finding buyers who actually convert is where every founder ends up rebuilding the wheel. especially when you're being smart about ICP and not just spraying volume.

[Metix Dev Platform @MetixDev]

The intent signal angle is underrated — finding businesses "ready to pay" only works if the buying signal is fresh. Most of the leads we see get filtered by recency (last 30 days job change / last 90 days funding) before they ever hit a sequence. We've been doing the same kind of filter upstream on the people-data layer: 900M profiles + 90M jobs + 30M companies, with the recency baked in. The hard part is keeping the source list fresh enough that the intent stays valid.

[Metix Dev Platform @MetixDev]

The maps angle is underrated — that's the one source no one's burned down yet. The piece that bit us hardest was the contact enrichment layer downstream of it: phone + website into a name + role + verified email is where scraping pipelines quietly rot. We ended up paying for a real people layer (900M profiles behind one key + MCP) just so the agent could ask in plain English instead of maintaining another enrichment script. Cuts the cleanup, but the maps sourcing part of your stack is genuinely the right move.

[Metix Dev Platform @MetixDev]

This is the cleanest split we've seen too — when the screening agent has a verified people layer behind it, the LLM stays in 'evaluate' mode instead of inventing candidate data on the fly. The part we're still chewing on is identity checks: in-interview signals are great, but they get a lot more useful when there's an external source of truth to match against (jobs they actually held, companies that exist, a 900M-profile record to anchor on). Curious where you're landing on that.

[Metix Dev Platform @MetixDev]

Pattern we see on data APIs too — give an agent a 900M-profile endpoint with no per-call logging, and it's getting scraped for spam within a week. Boring fix is logging the actual payload alongside the response, so you can tell a normal sourcing call from a list-dump before it leaves.

[Metix Dev Platform @MetixDev]

@kumarumt True on the MCP side too. The data layer is where it bites hardest — wrong people data + a loose agent filter and you've spammed 1,000 profiles in seconds. Same ladder, plus logging every tool call (who/what/when) before the agent touches real profiles.

[Metix Dev Platform @MetixDev]

@YashManghnani1 22 paying users with real behavior beats 1,000 cold signups every time. The behavior layer is also what we keep chasing on the people-data side — the records are easy, getting an agent to ask the right behavioral questions without breaking is where it actually gets interesting.

[Metix Dev Platform @MetixDev]

Exactly this. Most AI SDR embarrassment comes from the data layer, not the model — agents scraping the web end up with bad titles, stale companies, mismatched ICPs. We've been building that piece: 900M profiles / 90M jobs / 30M companies behind one key + MCP, so the agent asks in plain English and gets verified data before sending. Free 100 credits if you want to drop it into Ghost as a quality layer.

[Metix Dev Platform @MetixDev]

Depends on the job. For agents that enrich profiles or companies, staying power matters way more — the bottleneck is usually the data layer, not the model. We've been building that piece: 900M profiles / 90M jobs / 30M companies behind one key + MCP, so agents keep running without babysitting every query. Free 100 credits if you ever want to swap in a real people/company source.

[Metix Dev Platform @MetixDev]

This is exactly where AI SDRs are heading. We've been building the missing data layer for this stack: 900M profiles / 90M jobs / 30M companies behind one key + MCP, so the agent asks in plain English instead of scraping. If Signals ever wants a verified-people upgrade, the free 100 credits cover an eval pass.

[Metix Dev Platform @MetixDev]

@HNothing60052 do you need free api for this ？

[Metix Dev Platform @MetixDev]

不起眼的角落，M 3.1来了
minimax. yes❤️

[Metix Dev Platform @MetixDev]

This is the right instinct — data validation is what separates AI recruiter demos from AI recruiter production. We've been solving the same problem on the source side: 900M profiles / 90M jobs / 30M companies behind a 1-key API + MCP. If you ever want to swap the candidate lookup step for one that takes plain English queries, the free 100 credits cover an eval pass. Keep posting the day numbers.