AI use was distributed across individual tools and subscriptions without a unified operating model.
All case studies
Commercial case studyOngoing engagement
Theiagen GenomicsPathogen genomics and public-health bioinformatics
Building an AI-Enabled Operating Layer for a Genomics Company
A six-week foundation and continuing operator engagement spanning website control, quote-to-cash workflows, project governance, content operations, and AI adoption.
GenomicsAI OperationsWebsiteQuote-to-CashGovernance
01 / Original problem
The operational friction before the build.
Commercial, project, and delivery context lived across Notion, Google Drive, meetings, contracts, quotes, and individual expertise.
The company needed faster execution without bypassing scientific, healthcare, or client-review obligations.
02 / Exact deliverables
What was actually designed and delivered.
Website control and publishing path
A client-controlled staging/production workflow with natural-language editing support and review gates.
Quote-to-cash and SOW workflows
Structured, human-reviewed generation and CRM-linked workflow design with deterministic pricing safeguards.
Executive delivery board
A restricted outcome board with explicit In Review acceptance gates, resource links, and conservative ownership/status handling.
Content and AI governance
Two-tier review for routine marketing versus scientific/healthcare claims, plus role-aware AI enablement and subscription-audit planning.
03 / Architecture
Governed AI operations layer
Existing systems remain authoritative. AI helps connect, draft, and summarize; business owners and scientific reviewers retain approval.
01
Notion, Drive, meetings, and CRM
02
Governed context and workflow services
03
AI-assisted drafting and analysis
04
Business or scientific review gate
05
Approved system update
06
Outcome and acceptance tracking
04 / Implementation timeline
Six-week foundation, then continuing operations
The initial engagement moved from discovery into launch and handoff; later work focused on recurring governance, testing, acceptance, and operator enablement.
01–02
Map and prioritize
Inventory operating sources, risks, workflows, and near-term adoption targets.
03–06
Build and hand off
Advance the website, commercial workflows, delivery visibility, and repeatable editing process.
Ongoing
Govern and improve
Run acceptance gates, test workflows, train operators, and separate scientific claims from routine content.
05 / Before and after
Measurements with evidence labels.
Measured facts, observed workflow changes, evaluation targets, and missing baselines are kept distinct.
Delivery visibility
MeasuredBefore
Work and evidence spread across systems and meetings
After
32-card governed delivery board; 31 cards tied to a retained artifact or source record
Board state captured in the July delivery-governance review. An evidence-backed card is an outcome record linked to a source artifact; it is not automatically complete or client-accepted.
Website editing
ObservedBefore
Vendor and specialist coordination
After
Natural-language change request → commit → staged deployment workflow
Quote-to-cash
ObservedBefore
Manual and fragmented handoffs
After
Structured beta workflows with deterministic pricing and human review
Business KPI impact
Not instrumentedBefore
No locked baseline published
After
Client testing and acceptance still in progress
Methodology
How this case was substantiated
- Uses signed-engagement context, processed weekly operating reviews, and directly recorded deliverable state.
- Counts evidence-backed outcomes separately from completed or client-accepted work.
- Excludes private client workspace details and does not publish scientific or commercial claims without review.
Limitations
What the evidence does not prove
- Several workflows remain in beta, testing, review, or handoff; the case does not present them as fully adopted production truth.
- No revenue, labor-replacement, or scientific-performance outcome is attributed to the engagement.
- Internal customer data, contracts, pricing, account details, and private source inventories are omitted.
Map the operational friction before choosing the technology.
We can scope the workflow, evidence, adoption plan, security boundaries, and success measurements before recommending a build.
Related client work