For workforce boards and public agencies

See who's progressing, who's stalled, and why.

Each participant gets a job search that keeps running between appointments, by text on any phone. Case managers see what is blocking each person, and every match carries the reason it was made.

SAM.gov registered
UEI on file
SOC 2 certified

People reach it over
SMS
WhatsApp
Voice
Web

Overnight run · 300 participants

Live

Openings scanned tonight

1,204,331

Read every employer career page
1,204,331 openings
Dropped expired and duplicate listings
38,406 removed
Matched 300 participants on work they've done
2,914 candidates
Checked shifts, bus routes, and licenses
812 kept
Drafted for case managers to approve
300 ready by 7 am

Marcus T. · Warehouse Associate
Day shift ends before his curfew
Kept
Luis R. · Forklift Operator
Needs a certification he doesn't hold yet
Ruled out
Dana W. · Line Cook
12 minutes on the Route 4 bus
Kept

SOUND FAMILIAR?

Why workforce systems miss their numbers

If any of these sound familiar, the problem is the plumbing, not your staff.

Outputs, not outcomes

Programs count enrollments. Funders ask about employment, retention, and wage gain. The data that answers that question lives across four systems that don't talk to each other.

The handoff nobody owns

The moment a person moves between programs is the moment they disappear. Out of custody, off benefits, out of a dislocated worker event. One agency's responsibility has ended and the next one's has not started, and that gap is where the outcome actually gets decided.

One person, four case numbers

Corrections, the workforce board, the SNAP office and the unemployment system each hold part of the same person, and none of them holds the whole. Partner agencies serve the same participant without knowing it, funding pays twice for one outcome, and no single record can answer whether that person ended up employed.

Reporting consumes the staff you have

Quarterly performance reports get assembled by hand, by the same people who would otherwise be carrying a caseload.

The employer list nobody has

Staff fall back on the same short list of employers they already know. Which employers will actually consider this person, with this history, at this wage, reachable on the bus line they actually ride, is not in any system they have.

What if one record followed the person through every handoff, whatever program they came in through?

How it works inside your program

From intake to the performance report, on the data your funders will ask about

The program lifecycle, instrumented end to end

Phase 1

Enrollment where the person already is

Start the record before the handoff, not after it

Participants enroll and build a verified profile wherever the program already reaches them. Inside a facility before release, at the unemployment claim, at the benefits office, at the dislocated worker event. By phone, over SMS or WhatsApp, with no app to install and no login to remember. Skills, credentials and work history are structured on the way in and coded to standard occupation and skill codes automatically, so the record is comparable from day one and does not get retyped into a second system later. When one program's responsibility ends, the record does not restart. It follows the person.

Key Metrics

Enrollment at the point of contact • SMS and WhatsApp intake • Coded to SOC and O*NET on entry • One record across every handoff

Phase 2

Barrier identification

Spot the drop-off before it happens

The system watches engagement against your program's own thresholds and surfaces participants who go quiet, stall, or fall behind - while there is still time to intervene. Case managers get a ranked list of who needs them this week instead of discovering an exit at the quarterly review.

Key Metrics

Configurable at-risk thresholds • Early warning alerts • Intervention tracking

Phase 3

Matching and placement

Connect participants to real openings

Every participant is matched against 50M+ verified job postings pulled from the systems employers actually post into, with the specific skill or credential gap named for each role. Training recommendations follow from real regional employer demand rather than last year's plan.

Key Metrics

50M+ verified openings • Named skill gaps per role • Regional demand signals

Phase 4

Performance reporting

Prove it to your funders

Placement rates, time-to-placement, wage gain, and retention tracked automatically as they happen. Participant records map to the PIRL (ETA-9172) data elements your state already files, so the quarterly and annual lift stops being a manual reconciliation - and every figure carries the evidence behind it.

Key Metrics

PIRL-aligned records • ETA-9169 and WIPS ready • Evidence attached to every figure

EVERY NIGHT

Matched against dated evidence, not a generic profile.

Each participant's evidence file is scored against every open posting in the region with their real constraints: schedule, commute, credentials and program requirements. The employer's own hiring record is part of the score.

1
Evidence file
Denise's, 17 dated records
- ·Shift Lead, Midland Distribution, 2019 to 2026 · cycle counts, trained 12
- ·Forklift certification, county workforce center, Apr 2026
- ·OSHA 10, county workforce center, Apr 2026
- ·Inventory Clerk, Harbor Foods, 2016 to 2019
- ·+13 more, each dated and sourced
2
Filters
days · 45 min by bus · near her last wage
- ·Days only, clear of her Thursday program appointment
- ·Within 45 minutes by bus from Eastside
- ·No commercial license required
- ·$25 an hour or more, near what she was earning
3
Ranking
duties against evidence, employer file
- ·Duties on her evidence · 40 points
- ·Employer hiring record · 30 points
- ·Schedule and program · 20 points
- ·Commute · 10 points
4
Why it fits
in plain words, by text
- ·One line per match, written for a text
- ·Names the evidence it rests on
- ·Names the employer's record

Open jobs in the region96,210 open jobs

Warehouse AssociateEastside
Line CookRiverside
Delivery DriverDowntown
CustodianNorth End
Forklift OperatorSouthgate
Hotel StewardHarbor District
Security OfficerAirport
Prep CookEastside
Materials HandlerRiverside
Grounds CrewDowntown
DishwasherNorth End
Inventory SpecialistSouthgate
PorterHarbor District
Shipping ClerkAirport
Building MaintenanceEastside
Kitchen LeadRiverside
Warehouse AssociateDowntown
Line CookNorth End
Delivery DriverSouthgate
CustodianHarbor District
Forklift OperatorAirport
Hotel StewardEastside
Security OfficerRiverside
Prep CookDowntown
Materials HandlerNorth End
Grounds CrewSouthgate
DishwasherHarbor District
Inventory SpecialistAirport
PorterEastside
Shipping ClerkRiverside
Building MaintenanceDowntown
Kitchen LeadNorth End

Match · in the order they were postedFit /100

1
Warehouse SupervisorNorthwind Logistics
passes filtersForklift, crew lead and cycle counts all on her evidence. Hires from this program on file.
0
2
Yard AssociateNorthwind Logistics
passes filtersSame employer, but no lead duties and a 50-minute commute.
0
3
Shipping SupervisorCedar Health
passes filtersInventory counts and a crew of 12 on her evidence. Route 7, 18 minutes.
0
4
Inventory Control LeadSummit Supply
passes filtersThree years as inventory clerk plus seven as lead. Days, weekends off.
0
5
Receiving Clerk, nightsSummit Supply
passes filtersNights conflict with her Thursday appointment.
0
6
Inventory SpecialistRiverbend Market
passes filtersCounts match, but the pay is under her last wage.
0
7
Grounds CrewLakeshore Parks
passes filtersOutdoor crew work she has not done. Two buses each way.
0

Fit85 and up70 to 8455 to 69under 55·Built fromDuties on her evidence · 40Employer hiring record · 30Schedule and program · 20Commute · 10

Illustrative participant, employers and figures.

WHY THIS IS DIFFERENT

The labor market data your programs run on is broken. We rebuilt the pipe.

Public workforce programs are asked to place people using data that is months old, roughly a fifth fictional, and impossible to trace to a source. Metaintro is built on the opposite premise. A person leaving custody or coming off a benefit spell often has no laptop, no home internet, no active email address, and a phone number that will change twice this year. A portal that requires a login is a program that does not reach its population.

The numbers everyone is planning against

Federal payroll statistics are revised long after the fact - the preliminary benchmark revision published in September 2025 cut 911,000 jobs from the year through March 2025, the largest on record. Roughly 21% of online job postings are ghosts, never filled or already closed. And the largest labor data vendors sell billions of postings with no field-level record of where any of them came from.

We pull from the source, not from a resold feed

Most workforce data is licensed as an aggregate feed that has been deduplicated and resold several times over, which is exactly how ghost postings get in. Metaintro's ingestion pipeline pulls directly from the applicant tracking systems employers actually post into, through a pull-based streaming pipeline with backpressure and removal supervision. When a job closes, it leaves our index - so a participant is never sent to apply for something that no longer exists.

Matching on what the work is, not what it's called

Legacy labor exchange systems run keyword and job-title queries against a job bank. That is why a participant who worked the line at a poultry plant never surfaces for “quality control technician,” and why case managers stop trusting the match list. Metaintro runs hybrid retrieval - dense vector similarity combined with sparse lexical expansion over a vector index - so matching happens on the substance of the work and the skills behind it. The output is a ranked set of real openings with the specific missing skill or credential named for each one.

Every field carries its own receipt

Our data layer stores each fact with its source, its capture time, a freshness commitment, and a confidence score, re-derivable from an archived snapshot of the page as it was served. Anyone can crawl the web. Almost nobody can prove where a number came from. For an agency defending a training investment to an auditor or a legislature, that record is the difference between an assertion and evidence.

EU AI Act, Article 10 (data governance)

Built on international standards, crosswalked to yours

Occupations coded to ISCO-08 with crosswalks to SOC and O*NET, industry to ISIC with crosswalks to NAICS, skills to ESCO, geography to ISO 3166. Our data joins to what you already report on instead of forcing your analysts into a proprietary taxonomy only one vendor can read.

Change measured in hours, not quarters

The commitment we hold ourselves to publicly is median change-detection latency, in hours, against the multi-month lag of federal statistics. When a regional employer starts hiring, your program should know that week, not after the next revision.

Work that continues between appointments

A case manager carrying 300 participants can give each one about twenty minutes a month. Every other system in this category waits for that appointment. Metaintro assigns each participant a standing set of agents that keep working in between - scanning new postings, re-checking eligibility, flagging what changed - so the next conversation starts with what happened instead of catching up.

Recommendations that had to survive an argument

Most platforms return a match score. Ours get argued first. A separate model is assigned to refute each candidate recommendation, and a single editor agent is the only one that can admit it. What reaches your case manager is what survived a challenge, not what ranked highest.

An audit trail, not a black box

Every admission and every rejection is written to an append-only ledger with the evidence attached. When a legislator, auditor, or grant monitor asks why a participant was steered toward one pathway and not another, you can show them the record.

The whole labor market, not just your network

Referral platforms can only see the partner agencies inside their own network. Metaintro matches participants against 50M+ verified job postings nationwide, and names the specific skill or credential standing between a participant and each role.

SECURITY & COMPLIANCE

Government-grade security built in

SOC 2 Certified

Independently audited

Certified

Encrypted End to End

At rest and in transit

Full Audit Logging

Every action tracked

Role-Based Access

Granular permission control

SSO Ready

SAML, OAuth, Azure AD

WIOA Reporting

Federal reporting standards

SMS and WhatsApp

No app, no login required

Supports state and federal workforce reporting workflows · Data export available at any time · You own your data

NO RIP AND REPLACE

Designed to sit alongside your system of record

You are not replacing the system you file from. Metaintro is the intelligence layer above it, reading from the case management system you already run and writing back the matches, alerts and outcomes it cannot produce on its own. The systems below are the ones we see most often in state and county stacks. They are not a limit. If your department runs something else, including a system built in-house or one nobody outside your agency has heard of, we integrate to it.

SOC 2 Certified

SAM.gov Registered

PIRL Aligned

E-Verify Enrolled

State Systems of Record

The platform your agency files from

Geographic Solutions VOS
AJLA-TS
MonsterWorks
Launchpad
myOneFlow

Provider Case Management

What your subrecipients and CBOs run

Bonterra ETO
Bonterra Apricot
CaseWorthy
Casebook
Salesforce Public SectorNew

Reporting & Analytics

PIRL, WIPS and your own BI stack

WIPS (ETA-9172)
Power BI
Tableau
Azure Gov
AWS GovCloud

However your stack is actually built

Custom, legacy and in-house systems

REST and JSON APIs
Secure file transfer
Scheduled flat file exchange
SAML and OAuth SSO
Custom and in-house systems

Don't see your system? We support custom integrations via REST API and webhooks.

PROGRAM COVERAGE

Every program you run to put someone back to work

Most agencies run several funding streams over the same population, and the person moving between them is usually the same person. Metaintro treats a co-enrolled participant as one record rather than one record per program.

WIOA Title I and the labor exchange

Adult, Dislocated Worker, and Youth. Participant intake, barrier tracking, matching against real openings, and outcomes that map to the PIRL data elements your state already files.

RESEA and reemployment

RESEA carries a federal evidence requirement, which means the intervention has to be defensible and not only delivered. Our append-only evidence ledger was built for exactly that review - every recommendation, and the reason it was admitted, retained and linkable.

Dislocated workers and rapid response

A plant closure does not send you one participant, it sends you a cohort, and the clock starts the day the notice goes out rather than the day anyone walks in. Metaintro enrolls a WARN-affected group at once, matches each person against openings that actually pay near what they were earning, and shows the whole cohort's movement in one view instead of one file at a time.

SNAP E&T and TANF

Work participation and engagement requirements over a population that is almost always co-enrolled somewhere else, and that loses benefits when an hour goes unreported. Engagement is captured as it happens, in the channel the participant already uses, instead of reconstructed from a sign-in sheet at the end of the quarter.

Reentry and second chance

Reentry programs run before release, at release, and after, and ownership usually changes hands at the gate. Metaintro carries one continuous record across that handoff, matches against employers who will actually consider a record rather than employers whose postings merely imply it, and surfaces the incentives that move a hiring manager, WOTC and Federal Bonding included. Because the posting is unreliable evidence, we hold a dated employer evidence file instead of trusting the ad.

State prison and county jail are not the same program

A state prison population typically serves longer sentences, often with a parole plan and a known release date, which leaves room to build a plan months out. A county jail population is mostly pretrial, may leave in days, and may leave with no notice at all. The first needs a pathway. The second needs the record to already exist before anyone knows the release date. Metaintro runs both without forcing one into the other's template.

Veterans and JVSG

Jobs for Veterans State Grants funds DVOP and LVER staff against a population that carries priority of service by statute and a service record that civilian systems read poorly. Metaintro flags eligible participants for priority of service on intake, keeps the veteran-specific reporting separate from the general caseload, and gives specialists the same continuous record every other program line gets.

Vocational Rehabilitation

Longer engagements, richer accommodation and support needs, and outcome measures on their own schedule. Thresholds, milestones, and reporting are configured to the VR program rather than bent from a WIOA template.

Program operators and subrecipients

If you run centers or programs under contract - as an operator, a community-based organization, or a training provider - you can deploy across your own caseload without waiting on a statewide procurement, and report back to the funding agency in the format it expects.

Their career questions, answered

Give your program participants personalized insights powered by AI analysis of their resume and real-time market data

$124k

market rate

+12%

just type or talk

Build me a fresh resume

Am I underpaid?

Is my industry growing?

Is my job title at risk?

What jobs do I match to?

What skill adds the most value?

Build me a fresh resume

Am I underpaid?

Is my industry growing?

$124k

market rate

+12%

just type or talk

Is my job title at risk?

What jobs do I match to?

What skill adds the most value?

SIDE BY SIDE

Where Metaintro sits

Your system of record keeps the file. Your data vendor sells the market. Neither one works a caseload. That gap is what we fill.

Capability

State system of record

Labor market data providers

System of record for WIOA case files

Metaintro

Reads and writes

State system of record

Yes

Labor market data providers

System of record for WIOA case files

Reads and writes

Yes

PIRL and ETA-9169 reporting

Metaintro

Feeds it

State system of record

Yes

Labor market data providers

PIRL and ETA-9169 reporting

Feeds it

Yes

Real-time verified job openings

Metaintro

Yes

State system of record

State job bank only

Labor market data providers

Aggregate only

Real-time verified job openings

Yes

State job bank only

Aggregate only

Participant-level matching with named skill gaps

Metaintro

Yes

State system of record

Labor market data providers

Participant-level matching with named skill gaps

Yes

Barrier and at-risk alerts

Metaintro

Yes

State system of record

Manual

Labor market data providers

Barrier and at-risk alerts

Yes

Manual

Field-level source trail on every record

Metaintro

Yes

State system of record

Labor market data providers

Field-level source trail on every record

Yes

Evidence trail behind every recommendation

Metaintro

Yes

State system of record

Labor market data providers

Evidence trail behind every recommendation

Yes

Time to live

Metaintro

2-4 weeks

State system of record

6-18 months

Labor market data providers

n/a

Time to live

2-4 weeks

6-18 months

n/a

$8.5B

Annual federal and state workforce development budget requires measurable ROI

65%

Reduction in average unemployment benefit duration for participants

89%

Of program directors report improved stakeholder confidence with our data

50M+

Real-time job openings most government systems can't see

TRUSTED BY

Built for public procurement

State workforce agencies

Local workforce development boards

Vocational rehabilitation agencies

Departments of correction and reentry programs

K-12 districts and CCMR programs

State LMI and research offices

Program operators and subrecipients

SAM.gov

Registered, active UEI on file

SOC 2

Certified, independently audited

E-Verify

Enrolled employer

2022

Delaware C-Corp

Questions we hear from every workforce agency

If your question isn't here, book a call and we'll answer it live.

Real-time hiring data from every employer in your region

See your whole caseload — participant by participant

Book a walkthrough with our government team. We will show you the intake record, the early intervention queue, and the outcome report against your own program lines - and how it sits alongside the system you already file from.

SOC 2 certified. Encrypted at rest and in transit, with full audit logging. You own your data and can export it at any time.

Enterprise

AI career intelligence for universities and colleges

Boost career outcomes for accreditation reporting
Integrates with Banner, Workday & existing SIS
Real-time labor market data for 2,000+ occupations
SOC 2 certified & FERPA compliant

Learn more

Outplacement

Modern outplacement at a fraction of traditional cost

Deploy in 24 hours—no implementation delays
83% placement rate, 65% faster than industry avg
1/5 the cost of traditional outplacement firms
WARN Act compliance & audit-ready reporting

Learn more

Market Data

Real-time labor market intelligence API

50M+ jobs indexed with daily refresh cycles
RESTful API with 99.9% uptime SLA
Custom schemas & enterprise data delivery
SOC 2 Certified & GDPR compliant infrastructure

Learn more

See who's progressing, who's stalled, and why.

Why workforce systems miss their numbers

Outputs, not outcomes

The handoff nobody owns

One person, four case numbers

Reporting consumes the staff you have

The employer list nobody has

How it works inside your program

Enrollment where the person already is

Barrier identification

Matching and placement

Performance reporting

Matched against dated evidence, not a generic profile.

The labor market data your programs run on is broken. We rebuilt the pipe.

The numbers everyone is planning against

We pull from the source, not from a resold feed

Matching on what the work is, not what it's called

Every field carries its own receipt

Built on international standards, crosswalked to yours

Change measured in hours, not quarters

Work that continues between appointments

Recommendations that had to survive an argument

An audit trail, not a black box

The whole labor market, not just your network

Government-grade security built in

Designed to sit alongside your system of record

State Systems of Record

Provider Case Management

Reporting & Analytics

However your stack is actually built

Every program you run to put someone back to work

WIOA Title I and the labor exchange

RESEA and reemployment

Dislocated workers and rapid response

SNAP E&T and TANF

Reentry and second chance

State prison and county jail are not the same program

Veterans and JVSG

Vocational Rehabilitation

Program operators and subrecipients

Their career questions, answered

Where Metaintro sits

Built for public procurement

Questions we hear from every workforce agency

Who owns the data?

Do you have public sector experience?

Do we have to replace our case management system?

How does this affect our PIRL and WIPS reporting?

Does this work for programs beyond WIOA Title I?

We are a program operator, not the agency. Can we use this?

How quickly can we deploy?

Can you report in our state's format?

Can we customize thresholds and workflows?

How is participant information protected?

Real-time hiring data from every employer in your region

See your whole caseload — participant by participant

Explore Related Solutions

Enterprise

Outplacement

Market Data

Why workforce systems miss their numbers

Outputs, not outcomes

The handoff nobody owns

One person, four case numbers

Reporting consumes the staff you have

The employer list nobody has

How it works inside your program

Enrollment where the person already is

Participant Intake

Barrier identification

Early Intervention Queue

Matching and placement

AI Placement Engine

Performance reporting

Performance Reporting

Matched against dated evidence, not a generic profile.

The labor market data your programs run on is broken. We rebuilt the pipe.

The numbers everyone is planning against

We pull from the source, not from a resold feed

Matching on what the work is, not what it's called