Published writing
AI & agentic system design
Production runtimes, harness versus runtime boundaries, evals, guardrails, and integration patterns for multi-agent workflows.
Read the AI system design notes →Forward Deployed Engineeringfrom discovery to deployment
GI embeds engineers with your teams to discover workflows in production, commit to measurable outcomes, and ship secure agentic AI/ML and cloud systems your organization can deploy.
Discuss a projectForward Deployed Engineering (FDE)
That model: embedded engineers, measurable outcomes, and systems your teams can operate after deployment.
Three disciplines intersect in forward-deployed engineering (FDE): domain expertise (workflows and institutional context), software engineering (agents, integration, production systems), and business alignment (stakeholders, adoption, measurable outcomes). That intersection carries you from discovery through deployment: integration, evals, governance, and adoption from day one.
Services
Practice lanes that extend the FDE model above: depth across the stack. Pick a lane or combine engagements.
Machine learning & secure analytics
Model pipelines, adversarial robustness, and compliance-aware ML for regulated domains.
Learn more →High-performance & advanced computing
HPC integration, quantum-inspired methods, and specialized hardware roadmaps.
Learn more →Cloud & systems architecture
Container platforms, observability, and performance engineering on cloud infrastructure.
Learn more →Security & compliance
Post-quantum readiness, secure AI/ML, and architecture reviews for sensitive workloads.
Learn more →Practice focus
Digital intelligence & OSINT
Legal-grade open-source intelligence, attribution, and subpoena targeting for counsel.
OSINT practice landing →Featured writing
End-to-end LLM application design concepts
Draft chapter from the AI system design documentation: production runtime concerns, capability checklists, and reference material for agentic systems.
Case studies
Embedded engineering and agentic AI in production: GPU fleet remediation, pharma eval harnesses, secure ML platforms, and related FDE work. Court-tested digital intelligence for counsel lives on a separate practice track.
CS-01
Agentic GPU fleet remediation
Reduced GPU idle time and incident MTTR via SLURM-aware auto-remediation and observability integrated with agentic ops tooling.
CS-02
Pharma digital twins & synthetic persona evals
Built digital-twin and synthetic-persona evaluation flows with Langfuse-backed tracing for regulated pharma agent workflows.
CS-03
Letta buyer agent & custom eval platform
Shipped a Letta-based buyer agent with RAG over listing inventory, AWS deployment, and a custom eval platform for release gating.
CS-04
Why GRPO matters for LLM post-training
Fall 2025 research spanning PQC cryptographic discovery across 12 codebases and a GRPO security post-training proof-of-concept on Qwen3-4B, with a long-horizon agent framework design for compliance workflows.
All engineering case studies →OSINT & digital intelligence →
People
GI is led by practitioners with deep experience in quantum computing, secure AI, HPC, and regulated industry delivery. We embed senior engineers who communicate clearly with your teams and leadership.
Dr. Jason Larkin
Founder & Principal Consultant
Dr. Jason Larkin is the founder and principal consultant at GI Consulting, with over a decade of experience in quantum computing, secure AI, and high-performance computing for regulated industries.
He has delivered quantum advantage assessments and secure AI implementations across healthcare, pharmaceutical, defense, and financial sectors, embedding with your teams from discovery through production.
Prior to founding GI Consulting, he served as a technical consultant for major defense contractors and healthcare organizations, building quantum frameworks and HIPAA-aware AI/ML pipelines.
Full profile and contact →