Mission, in five years: Screening candidates

In March 2024, a major European service marketplace processed over 12,000 plumbing requests where the assigned provider carried a 4.9-star rating, yet 14% of these jobs resulted in secondary property damage or incomplete work. The metric failure was structural: five-star ratings on consumer platforms measure politeness and punctuality, not structural engineering competence or circuit topology knowledge.

Selecting labor for physical tasks—whether industrial wiring, structural masonry, or high-value art handling—remains stuck in a binary trap. Platforms either accept self-reported resumes and static PDFs, or they charge prohibitive fees to run one-off criminal background checks that say nothing about whether a technician can properly balance a heat pump.

The Five-Star Myth and the Paper Certificate

Legacy platforms treat screening as a gatekeeping event executed once at sign-up. A contractor uploads a scanned vocational diploma, passes an identity check, and receives a verified badge. From that point forward, the platform relies on post-hoc star ratings to monitor quality. This model incentivizes rating inflation; clients hesitate to ruin a worker's livelihood over a minor defect, while workers lobby for quick payouts before hidden defects emerge.

Paper certificates decay the moment they are issued. A master electrician certified in 2012 may have no documented experience with modern microinverters or ISO-compliant heat recovery systems. Conversely, a self-taught technician with five years of flawless complex installations on specialized machinery is often filtered out by automated keyword screeners looking for specific institutional titles.

Screening must move from static documentation to dynamic telemetry. The question is not what a candidate claimed to know during onboarding, but how their operational history across real-world tasks demonstrates current competence.

Real-Time Telemetry Over Resume Padding

WEVONE’s long-term architecture views screening not as an HR department’s triage step, but as a continuous computational query across independent operational ledgers. When a task poster requests an industrial equipment installer in the Mission universe, candidate matching does not begin with keyword matching against a self-written profile.

Instead, Mia—WEVONE’s core AI infrastructure—queries the candidate's cross-universe profile across multiple vectors. If the candidate regularly rents specialized diagnostic equipment through the Tools universe and returns it on schedule with clean calibration logs, that telemetry carries measurable weight. If their dispute log in Mission demonstrates zero escrow holds over twenty high-variance assignments, their reliability weight increases proportionally.

This approach relies on explicit WEVONE mechanics:

• Cross-Universe Activity Signals: Reliability in Tools (equipment care) and Pilote (transport precision) feeds directly into skill confidence scores in Mission.
• Transactional Escrow Telemetry: Successful escrow releases without post-dispute intervention form a immutable record of completed performance, distinct from subjective reviews.
• Mia’s Context Memory: Mia analyzes execution details—such as time-to-completion against job complexity parameters—rather than generic customer feedback comments.

By aggregating signals across discrete universes, screening becomes an emergent property of platform activity rather than a self-reported resume.

Scenario: The Midnight Electrical Emergency in Lyon

At 11:45 PM on a Tuesday, a mid-sized cold-storage facility outside Lyon experiences an main breaker failure. The plant manager posts an emergency job on Mission: immediate dispatch required for a technician certified on 400V industrial switchgear.

Under a legacy platform model, three contractors respond. The manager sees three profiles, all displaying 5-star ratings and identical claimed skill sets. The selection is a gamble based on photo professionalism or response speed.

Under WEVONE’s five-year screening vision, Mia executes a contextual graph check in under two seconds:

1. Candidate A holds a 2018 vocational diploma uploaded to the system. However, their last four Mission completions involved residential lighting, and they have no record of handling high-voltage diagnostic tools on the platform.
2. Candidate B lacks a traditional master's diploma but has completed eleven verified commercial power interventions in the past six months. Their transactional escrow ledger shows zero disputes, and their Tools history confirms recent usage of calibrated thermal imaging cameras.
3. Candidate C has high star ratings but an open dispute window from a job three days prior involving improper grounding on a similar panel.

Mia routes the priority offer directly to Candidate B, presenting the facility manager with verified operational data rather than star averages. The selection relies on objective operational traces rather than self-promotional text.

Current Reality vs. Five-Year Ambition

To evaluate this framework honestly, we must delineate what is operational today from what remains an engineering goal.

• Fact (Live Today): Basic identity verification, dispute tracking, and transactional escrow release mechanisms are fully functional in WEVONE's core architecture. Escrow funds unlock strictly based on milestone verification.
• In Beta: Cross-universe signal integration between Tools and Mission. Mia currently tracks tool rental durations and dispute histories, but full cross-universe score weightings are undergoing calibration to prevent unintended bias.
• Ambition (5-Year Vision): Zero-knowledge credential verification linked to European national databases, paired with predictive skill decay modeling. The goal is a system where local labor credentials require zero manual submission while offering complete cryptographic proof of operational competence.

The Edge Cases: Cold Starts and Privacy Boundaries

Dynamic screening carries explicit technical and ethical trade-offs. The cold-start problem is the most acute: how does a skilled artisan newly arriving in a city—or joining WEVONE for the first time—compete against established platform workers with deep ledger histories?

If the system over-indexes on historical platform telemetry, it risks creating closed local monopolies. WEVONE addresses this by designing explicit "probationary escrow paths." New entrants can accept tasks with shorter dispute windows or opt for peer-audited initial jobs where a high-reputation provider validates the work on-site, rapidly seeding their initial context memory without requiring months of low-paid labor.

Privacy limits present another bound. Continuous telemetry must never turn into pervasive surveillance. Workers maintain granular control over which cross-universe signals are exposed to potential clients. A worker may choose to hide their Pilote or Nest activity from their Mission profile; doing so simply removes those specific telemetry boosts from their confidence score without penalizing their base credential status.

Automating screening is not about replacing human judgment with opaque algorithmic gatekeeping. It is about removing the friction of paper verification and replacing superficial star ratings with verifiable physical performance records.