Sold by: Kriv AI
5-day private Claude Code workshop for healthcare and life-sciences engineering teams (2–10 engineers). Ship production AI workflows on Amazon Bedrock with HIPAA-aligned guardrails, MCP integration, evaluation harnesses in CI, and a senior-engineer review of every pattern — taught on your own codebase, not slides.
Curriculum covers Bedrock routing, permission modes, CLAUDE.md hierarchy, subagents + Skills + hooks, FHIR / HL7 / SMART-on-FHIR scaffolding, PHI redaction via Comprehend Medical or Microsoft Presidio, append-only audit trail patterns suitable for 21 CFR Part 11 review, golden-dataset evals in CI, and headless Claude Code in CodeBuild for nightly refactor-bot jobs. Delivered by an AWS Select Tier Services Partner and Anthropic Claude Partner Network member focused exclusively on regulated industries.
Overview
Train your healthcare engineering team to build with Claude Code the right way — fast, auditable, and aligned to your compliance bar.
Kriv AI's 5-day Claude Code Enterprise AI Training is a private, cohort-based engagement for regulated engineering teams (2–10 engineers) who need to move past demos and start shipping production AI workflows on AWS. The workshop combines instructor-led deep dives, pair-programming labs on your actual repository, and a senior-engineer review of every pattern before it leaves the room.
What's included — day by day
Day 1 — Foundations & Safe Defaults. Claude Code architecture, Bedrock routing (CLAUDE_CODE_USE_BEDROCK=1), model selection across Opus 4.7 (1M context), Sonnet 4.6, and Haiku 4.5. Permission modes (why bypassPermissions is banned in regulated envs), the CLAUDE.md hierarchy, cost observability, and secure credential handling.
Day 2 — Subagents, MCP, Hooks & Skills. Task/subagent patterns with per-agent model scoping, custom slash commands, Skills, and hook events (PreToolUse, PostToolUse, UserPromptSubmit). Build stdio and HTTP MCP servers, debug with MCP Inspector, and wire PHI-regex deny hooks that block egress to un-BAA'd endpoints.
Day 3 — Healthcare Integration Patterns. FHIR R4 resource modeling, HL7 v2 mapping, Epic/Cerner SMART-on-FHIR scaffolding, de-identification via Microsoft Presidio orchestrated by a Claude subagent, and synthetic test data from Synthea. Generate an append-only hash-chained audit trail suitable for 21 CFR Part 11 review.
Day 4 — Evals, Observability & Human-in-the-Loop. Golden datasets, deterministic evals, regression gates in CI, and operational dashboards. Teams leave with a working evaluation harness wired into their pipeline — not a slide.
Day 5 — Production Hardening & Handoff. Git-native workflow (signed commits, DCO, gh pr create), headless Claude Code in CodeBuild for nightly refactor-bot jobs, prompt-injection threat modeling on clinical data, rollout playbook (pilot → team → org), and a documented internal runbook your team owns.
Who it's for
VP / Director of Engineering, Head of AI Platform, and Principal Engineers at healthcare providers, payers, PBMs, digital health, health-tech, clinical research organizations, and medical-device companies (typically 200–5,000 employees, 40–400 developers). Ideal when your org has committed to Claude and needs a defensible internal practice — not another pilot.
Compliance posture
Curriculum maps to HIPAA Security Rule controls (45 CFR §164.308 administrative, §164.310 physical, §164.312 technical — including audit controls), 21 CFR Part 11 audit trail and ALCOA+ data integrity, and HITRUST CSF v11.2.0 AI Security controls. We cover documentation patterns that support Colorado SB 24-205 impact assessments (enforcement 30 June 2026) and Texas TRAIGA (HB 149) disclosure obligations (effective 1 Jan 2026). Kriv AI provides engineering methodology only — not legal, regulatory, clinical, or compliance-attestation advice.
What you leave with
A trained team; a configured Claude Code baseline committed to your repo; a working MCP integration; an evaluation harness running in CI; a PHI-redaction + audit-log hook suite; an organization-specific CLAUDE.md template; and a written internal playbook covering security, cost controls, and incident escalation. All artifacts are customer-owned.
About Kriv AI
Kriv AI is a US-based AI consulting firm focused exclusively on regulated industries — healthcare, life sciences, insurance, and financial services. Kriv AI is an AWS Select Tier Services Partner and a member of the Anthropic Claude Partner Network (approved April 2026).
Important — infrastructure costs
This listing covers training services only. Amazon Bedrock model invocation, AWS compute, storage, networking, and any third-party software costs incurred during and after the workshop are billed by AWS directly to the customer and are not included in the Kriv AI engagement fee. The customer is solely responsible for executing its own Business Associate Agreement (BAA) with AWS and for obtaining internal security approval for Claude Code use.
Get started
Contact info@kriv.ai or +1 732 433 5564 to scope a private offer. Most engagements kick off within 3–4 weeks of contract signature.
Highlights
- 5-day private workshop for 2–10 healthcare engineers, taught on your own codebase. Covers Claude Code architecture, Bedrock routing (CLAUDE_CODE_USE_BEDROCK=1), permission modes, model tiering across Opus 4.7 / Sonnet 4.6 / Haiku 4.5, subagents + Skills + hooks, MCP server development, FHIR / HL7 / SMART-on-FHIR scaffolding, PHI redaction via Comprehend Medical or Microsoft Presidio, golden-dataset evals in CI, and headless Claude Code in CodeBuild — production patterns, not demos.
- HIPAA-aligned engineering patterns with Bedrock-only egress, PHI-regex deny hooks, append-only hash-chained audit trail patterns suitable for 21 CFR Part 11 review, and a working evaluation harness wired into your CI pipeline by Day 4. Maps to HIPAA Security Rule §164.308/310/312/316, HITRUST CSF v11.2 AI Security, Colorado SB 24-205 impact-assessment posture (enforcement 30 Jun 2026), and Texas TRAIGA disclosure obligations (effective 1 Jan 2026).
- Delivered by Kriv AI — AWS Select Tier Services Partner and Anthropic Claude Partner Network member (approved April 2026) — focused exclusively on regulated industries (healthcare, life sciences, insurance, financial services). 60-day post-workshop follow-up office hours included. Engagement fees cover Kriv professional services only; customer pays AWS directly for any Bedrock and AWS infrastructure consumption. Methodology only — not legal or compliance-attestation advice.
Details
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Pricing
Custom pricing options
Pricing is based on your specific requirements and eligibility. To get a custom quote for your needs, request a private offer.
Legal
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Resources
Vendor resources
Support
Vendor support
Primary support contact: info@kriv.ai · +1 732 433 5564 · https://kriv.ai/support
Response SLA: Kriv AI responds to AWS Marketplace inquiries and post-private-offer kickoff requests within 2 business days during US business hours (Eastern Time, Monday–Friday). This meets and slightly improves on the AWS Professional Services 2-business-day minimum.
Engagement-specific escalations are routed to the assigned Kriv engagement lead within 1 business day on request. Customers receive a dedicated Microsoft Teams or Slack channel with named engagement lead at kickoff for the duration of the workshop and the 60-day follow-up window.
Hours of operation: Monday–Friday 9:00 AM – 6:00 PM Eastern Time (US). Off-hours messages acknowledged the next business day. Critical-incident escalation contact provided in the kickoff packet.