Methodology
Five operating principles. One AI/Expert Reconciliation Log.
How HSG translates Maurice House's and Kevin Latner's 38- and 25-year FAS careers into a USDA-maintainable operating playbook within a twelve-month period of performance — and the AI-augmented delivery methodology that makes the eight-guide cadence feasible.
Five Operating Principles
How HSG approaches every PWS scope area.
Inside-out, not outside-in
Process guides are written from the USDA FAS operational vantage point rather than imported from generic humanitarian-programming frameworks. Maurice House and Kevin Latner contribute that institutional voice directly — the documents read like they came from inside the agency that has to run them.
AI-augmented, expert-reconciled
HSG's AI-augmented service delivery methodology accelerates the mechanical 70–80% of process-documentation work (research synthesis, draft generation, formatting, cross-reference checks) while preserving senior judgment on every consequential recommendation. Every AI-suggested output is paired with an expert reviewer disposition in HSG's AI/Expert Reconciliation Log.
Title II + non-emergency unified view
Although USAID-era Food for Peace was operationally bifurcated between emergency Title II and non-emergency programming, HSG's process guides treat them as a single integrated operating envelope so USDA staff can flex resources across emergency and non-emergency states without process discontinuity.
Cargo Preference + domestic manufacturing integrated
PWS 3.2 explicitly calls for triangulation of Cargo Preference compliance, U.S. domestic specialty-formulated commodity manufacturing capacity, and emergency response timing. HSG treats all three as one decision space — a single trade-off framework rather than three separate analyses USDA staff have to reconcile.
Risk-tiered country selection
PWS 3.4 calls for prioritization that considers more than just finances. HSG builds the country-selection methodology around a transparent multi-factor weighted score that USDA can audit, reproduce, and update annually — defensible in Congressional testimony and OIG response.
Bilateral co-investment deal architecture
The administration's America First foreign-aid doctrine — operationalized in the State Department's 31 bilateral Global Health Strategy MOUs ($20.6B total, $7.8B recipient-country co-investment, glide-path declining U.S. share) — is the template HSG applies to FFP process-guide design. Country Selection becomes deal-readiness assessment. Communications becomes value-chain articulation. Geography & Commodity Prioritization adds a counterpart-contribution viability dimension. Anchored by Audrey McGuire's Wall Street capital-markets background and Kevin Latner's development-to-business transition consulting practice.
AI-Augmented Delivery
The 80/20 split that makes the eight-guide cadence feasible.
Federal advisory work has two layers — the mechanical 70–80% (research, drafting, formatting) and the judgment 20% (interpretation, recommendation, audit defense). AI without expert review is reckless on the 20%. Expert work without AI on the 80% is wasteful in a twelve-month firm-fixed-price contract. HSG's methodology lets the LLM draft the 80% and senior FAS experts confirm every consequential action.
AI Workflow Build
LLM (Anthropic Claude API or Azure OpenAI in federal-cloud-eligible regions) drafts the mechanical 70–80% of each process guide — research synthesis from authoritative sources, draft narrative generation, formatting consistency, cross-reference resolution. Expert reviewer confirms or edits every consequential output.
AI/Expert Reconciliation Log
HSG's signature transparency deliverable. Every AI-suggested action is paired with the expert reviewer's disposition, timestamp, and underlying input lineage. The Log is the audit-defense artifact for any post-engagement OIG, GAO, or Inspector General review.
Federal AI Governance
Aligned to OMB M-25-21 (Accelerating Federal Use of AI), M-25-22 (Driving Efficient Acquisition of AI in Government), Executive Order 14179, and the NIST AI Risk Management Framework. Documented model cards, use-case registration, and human-on-the-loop discipline throughout.
In Practice
This methodology is not theoretical.
HSG has applied the AI-augmented delivery methodology on the USDA APHIS overseas workforce engagement (Mexico, Panama, Guatemala — rated Exceptional across all categories on signed Past Performance Questionnaire), on the VA Lease Contract Administration engagement (58 active VHA leases / 2.5M sq ft / 30%+ administrative time reduction), and on the Syngenta GMO Corn expert-witness engagement (~$500M settlement) where Jelani House synthesized U.S., Chinese, and EU regulatory frameworks.
On this contract, the methodology compresses the eight process-guide deliverables into a cadence that fits the twelve-month POP — and produces an AI/Expert Reconciliation Log that becomes the audit-defense artifact for USDA's post-engagement use.