FieldPact

FieldPact is a self-described workforce platform for skilled trades that structures project needs into "Crew Specs" and "Crew Packs", treating the deployable crew as the unit of work rather than individual applicants. The system uses GPT-5.6 for language reasoning and structured drafting, with deterministic TypeScript handling credential validation, role arithmetic, and assignment logic. It is presented as a demo for OpenAI Build Week, built with Next.js, React, and TypeScript.

The description states that FieldPact transforms a plain-language project brief into a validated Crew Spec and three Crew Pack options, with transparent operational lenses for contractor selection. It includes features like same-slot recovery when a worker declines, worker-controlled Career Proof, and recruiter Copilot for trade-specific screening without issuing hire/reject verdicts.

The author claims the platform addresses two core problems: finding complete crews rather than individual workers, and making successful work portable as "Career Proof". However, no evidence of revenue, customers, or adoption is provided. The system is described as a demonstration with fictional data and no real-world staffing outcomes.

The description states that FieldPact "transforms a contractor’s plain-language project brief into a validated Crew Spec and three complete Crew Pack options". It uses GPT-5.6 for language reasoning and structured drafting, while deterministic TypeScript handles credential validation, role arithmetic, and assignment logic.

The system is described as having:

• A "Crew Spec" that covers project details, required roles, qualifications, logistics, terms, assumptions, and unknowns
• Three "Crew Pack" options with transparent operational lenses (evidence depth, lower mobilization exposure, stronger ready-alternate coverage)
• A workflow that includes contractor submission, Crew Spec confirmation, deterministic validation, Crew Pack review, offer preparation, worker response, same-slot recovery, and Career Proof generation

The platform is built as a responsive Next.js, React, and TypeScript application with server-only AI routes using OpenAI Responses API with GPT-5.6 Structured Outputs.

The description states that FieldPact was designed around two core ideas:

1. The contractor should be able to evaluate a complete crew, not a pile of disconnected applicants
2. Successful work should compound into worker-controlled Career Proof

The platform positions itself as different from most staffing products by treating the deployable crew as the operational unit rather than individual applicants. It describes its workflow as: "project brief → confirmed Crew Spec → deterministic hard gates → complete Crew Pack options → human selection → transparent offers → worker response → same-slot recovery → worker-controlled Career Proof".

The authors claim this approach addresses a problem where conventional staffing systems are organized around applicants while field operations are organized around crews. They also state that successful work often disappears once a project ends, and FieldPact was designed to make accomplishments portable.

The description states that FieldPact targets contractors who need to find complete crews for skilled trades projects. It specifically mentions "contractors" and "staffing recruiters" as key personas, with the public need being substantial based on U.S. Bureau of Labor Statistics projections about 81,000 electrician openings per year from 2024–2034.

The platform is described as addressing a problem that affects skilled trades staffing where contractors can often find individual workers but struggle to form complete crews at the time they are needed and at workable cost. The authors note that "the Associated General Contractors of America continues to report difficulty filling hourly craft roles."

However, no specific customer segments or personas beyond "contractors" and "recruiters" are detailed in the description.

Not evidenced. The description does not contain any information about pricing models, revenue streams, or business model details.

The platform is built with:

• Next.js, React, TypeScript
• Server-only AI routes using OpenAI Responses API with GPT-5.6 Structured Outputs
• Deterministic TypeScript for credential validation, role arithmetic, and assignment logic
• Shared Zod contracts connecting server routes, fixtures, validators, tests, and UI state
• CSS Modules for styling
• Three.js for visualizations (with SVG fallback)
• Playwright for end-to-end testing
• Vitest for unit testing

The system uses GPT-5.6 for language-intensive parts including transforming project briefs into Crew Specs, generating screening questions, organizing evidence, and drafting Career Proof. Deterministic TypeScript handles credential expiry, availability compatibility, role arithmetic, unique assignment, pack generation, declined-worker exclusion, prohibited topics, source-ID checks, and unsupported numeric claims.

The demo uses fictional profiles and no real personal data. Fixture mode is the default judging path with zero-cost operation, while live GPT-5.6 mode is available as optional paid opt-in.

Not evidenced. The description states that this is a demonstration for OpenAI Build Week and that broader structured validation with contractors, staffing recruiters, and tradespeople is still ahead. No evidence of revenue, customers, or adoption is provided beyond the authors' own claims about the project's scope.

The description states that most staffing products focus on applicant matching, shift filling, or individual vetting, while FieldPact treats the deployable crew as the operational unit. It claims this approach is different because it:

• Treats the crew rather than individual applicant as the unit of work
• Presents three transparent pack-level trade-off lenses
• Implements deterministic same-slot recovery when a worker declines
• Uses one evidence vocabulary across contractor, recruiter, and worker experiences
• Provides complete terms before commitment
• Offers source-grounded Career Proof controlled by the worker

However, no specific competitors or competitive positioning details are provided in the description.

1. Unproven commercial viability: The platform is described as a demo with fictional data and no real-world staffing outcomes or customer adoption.

1. Over-reliance on AI without clear human decision-making boundaries: While the description states that "accountable people make every employment decision", it's unclear how this is enforced in practice given extensive AI involvement in various workflow steps.

1. Limited scope: The demo focuses on one fictional scenario (ten-person electrical crew for a Dallas data-center project) and doesn't demonstrate broader applicability or scalability.

1. No evidence of real-world validation: The description explicitly states that broader structured validation with contractors, staffing recruiters, and tradespeople is still ahead, indicating no current market testing.

1. Technical complexity vs. practical implementation: The platform uses complex technical components (GPT-5.6, deterministic TypeScript, Three.js) but the demo appears to be a single workflow path rather than a production-ready system.

1. What specific market pain points have you identified in skilled trades staffing that led to this approach?
2. How do you plan to validate your "crew-as-unit" thesis with actual contractors and staffing teams?
3. What are the key operational challenges you expect to face when scaling beyond a single demo scenario?
4. How will you ensure accountability and human decision-making in an AI-heavy workflow?
5. What specific metrics or KPIs do you use to measure success of your "crew" approach versus traditional applicant-based hiring?
6. How do you plan to handle the transition from fictional data to real-world implementation with actual workers and contractors?
7. What are the key technical challenges you anticipate in implementing deterministic validation at scale?