Platform & AI engineering · former fintech CTO / acting CISO

I build AI platforms —and the teams that own them.

VP/CTO-level platform & AI engineer who ran trading infra at 99.999% uptime while AUM scaled to $15B. Now I architect multi-model AI and agentic systems in Rust — and ship them.

Available now · advisory, fractional & build — taking a few new engagements this quarter
Rust agentsLLM platformsMulti-model orchestrationBlockchain · SolanaSecurity & SOC 2AutomationAgentic pipelinesKubernetes · AWSMCP toolingFractional leadership

Outcomes I deliver

// in production, not slideware

The kind of results I bring to an engagement — measured, in production, under real constraints.

$0B
AUM the platform scaled to
0%
Uptime · trading & risk infra
−0%
Model deploy time
−0%
LLM inference cost
−0%
Onboarding time
−0%
CI/CD cycle time
NF
// who I am

I make systems work when the stakes are real.

For 13 years I was a fintech CTO and acting CISO — running trading and risk platforms at 99.999% uptime while the business scaled from ~$800M to $15B in assets. I built that engineering org from its first hire to ~40 people across time zones, and owned the security, IT, and disaster-recovery framework end to end.

Today I architect multi-model LLM and agentic AI systems in Rust — the platforms, the agent runtimes, and the governance around them — and ship them to production on Kubernetes and AWS. I still write the hard parts myself.

I care about the unglamorous things that decide whether AI actually ships: security, reliability, cost, and the team that has to own it after I'm gone.

How I build

// principles

The convictions I bring to every engagement — the reasons the work actually ships.

01

Production is the only proof

A demo that never ships is a liability. I optimize for the boring path to production — deploys, rollbacks, on-call — because that's where value actually lands.

02

Secure & auditable by default

I came up in regulated fintech as an acting CISO. AI systems get the same rigor: SSO, access control, per-team cost attribution, and an audit trail — not bolted on later.

03

Own the stack, then hand it back

I write the hard parts and set the standards, then leave teams a platform they can run without me. Leverage beats heroics.

04

Measure or it didn't happen

Deploy time, inference cost, MTTR, uptime. If we can't watch it move, we're guessing. Every engagement ships with the numbers that matter.

Work with me

// engagements

Need a second opinion, a fractional leader, or someone to build and ship it? Here's how to put 25 years to work.

Advisory

AI Platform & Agent Advisory

Architecture reviews, agent-system design, and a senior set of eyes on your AI roadmap before you bet the quarter on it. Fixed-scope sprints or an ongoing retainer.

  • Platform & agent-system architecture review
  • LLM orchestration & MCP infrastructure strategy
  • Security, SOC 2 & AI-governance review
  • Technical due diligence
Book a 30-min call
Fractional

Fractional VP Platform / AI

Senior platform & AI leadership without a full-time hire. I own the roadmap, set the standards, hire and level up the team — and still write the hard parts.

  • Platform & AI roadmap + standards
  • SRE, SOC 2, FinOps discipline
  • Hire, mentor, and unblock the team
  • A few focused days a month
Book a 30-min call
Build & ship

Build the system, to production

Hands-on engagements where I come in and ship it — the agent platform, the Rust service, the Kubernetes/AWS infra — production-grade and SOC 2-ready. Fixed scope, real delivery.

  • Agent runtimes & LLM platforms in Rust
  • Kubernetes / AWS / Terraform infra
  • Production-grade, regulated-ready
  • Clear scope, shipped
Book a 30-min call
Full-time

Open to the right full-time role

If it's genuinely the right fit, I'm open to VP / Staff / Principal AI-Platform roles too — someone who owns the platform strategy and writes the hard parts.

  • VP/CTO-level + still hands-on
  • AI platform & agent systems
  • 25 years across hard, regulated domains
  • Let's talk
Book a 30-min call
01

Selected Work

github.com/njfio

A few systems I designed, built, and shipped end-to-end — pick one and go deep.

// more work
· Web3 / Trading

On-Chain Trading & Blockchain Infra

Production blockchain across Solana, Ethereum, Polygon & Cosmos — an on-chain staking program for a sports-betting product, an order-management system and trade-signal pipelines for Polymarket, and a real-time market-insights analytics engine.

SolanaEthereumOMSRust
· AI Platform

Enterprise AI Governance

Secure, audited, fully-tracked LLM usage via Obot & Bifrost — centralized access control and per-team cost attribution across a ~50-engineer organization.

AI GovernanceSSOFinOpsObot/Bifrost
· Quant Infrastructure· NDA

Risk & Trading Platforms

Architected AWS/Azure infrastructure for quantitative analysis and risk modeling at an insurance-linked-securities firm — 99.999% uptime, SOC 2, FinOps cutting infra spend 35%, while AUM scaled from ~$800M to $15B.

AWSAzureRisk ModelingSOC 2
02

Open Source

github.com/njfio

~37 public repos — Rust, AI agents, and the plumbing in between.

03

Experience

2000 — present

Twenty-five years turning infrastructure into leverage — and teams into platforms.

2022 — Now
VP Platform Engineering, AI R&D
Multi-model LLM & agentic AI orchestration in production (deploy time −40%); MCP integrations cut client delivery ~30%. Shipped blockchain infra across Solana, Ethereum, Polygon & Cosmos — including an OMS and trade-signal pipelines for Polymarket. Stood up enterprise AI governance (Obot/Bifrost) across a ~50-engineer org. Led a platform team of up to 30 — uptime +15%, MTTR −40%, onboarding −50%, CI/CD −60%.
FP Complete Corp.
2013 — 2021
Chief Technology Officer & acting CISO
Built the global engineering org as first hire — recruited 60+ engineers to a steady-state ~40 across time zones — while the business scaled AUM from ~$800M to $15B. As acting CISO, owned the full security, IT & disaster-recovery framework; implemented SOC 2 and financial-regulatory compliance. AWS/Azure for quant & risk modeling at 99.999% uptime; FinOps cut infra spend 35%.
Nephila Advisors
2008 — 2012
Senior Systems Engineer
Ran a 1,000–2,000-node VMware estate for hedge-fund operations at 99.999% uptime; led virtualization and disaster-recovery strategy (hardware −40%). Built automated testing for all HPC and cluster designs, and managed six SAN fabrics plus the global LAN/WAN.
Nephila Capital
2007 — 2010
Enterprise Consultant · Bermuda
Enterprise infrastructure consultant based in Bermuda. My first client was Nephila — the start of a long run with the firm — serving them full-time across storage, disaster recovery, and virtualization.
Applied Computer Technologies
2000 — 2007
Support Engineer III & Escalation Engineer
Level-III support and escalation engineering for enterprise customers — owning the hardest storage, server, and system failures other tiers couldn't resolve, under SLA.
Hewlett-Packard
04

Capabilities

// the stack

The tools I reach for, grouped by where they earn their keep.

AI / LLM
  • Agentic AI & agents
  • LLM orchestration
  • Model Context Protocol
  • Multi-model fusion
  • Token / cost optimization
  • AI governance (Obot/Bifrost)
  • OpenAI · Anthropic · LangChain
Languages
  • Rust
  • Python
  • TypeScript
  • JavaScript
  • Go
  • SQL
  • Bash
  • PowerShell
Platform & Cloud
  • Kubernetes (EKS)
  • AWS · Azure
  • Terraform · Helm · Docker
  • GitHub Actions CI/CD
  • Prometheus · Grafana
  • OpenTelemetry
  • PostgreSQL · Redis
Security & Compliance
  • CISO-level leadership
  • Security / IT / DR frameworks
  • SOC 2 & financial-reg compliance
  • Identity & SSO (Okta/OAuth2)
  • Network security
Infrastructure
  • VMware (1,000+ node estates)
  • SAN & storage architecture
  • HPC & cluster design
  • LAN/WAN networking
  • DR & business continuity
Product & Web
  • Custom website & web-app design
  • Full-stack (Rust · Vue · TypeScript)
  • Streaming chat & gen-AI UIs
  • Front-end build & UX
  • Web performance
Automation
  • Workflow & marketing automation
  • Multi-agent GTM engines
  • Data scraping & analysis engines
  • Agentic & data pipelines
  • Rapid prototyping (15+ presales)
Domains
  • Trading systems & OMS
  • Regulated fintech (SOC 2)
  • Blockchain (Solana/Ethereum/Polygon/Cosmos)
  • Enterprise AI governance
  • Presales / solutions engineering
  • SRE & incident management
05

Range

beyond the stack

Proof I ship across domains: native apps, published fiction, original research.

Shipped iOS apps & games

iOS Game · Swift / SpriteKit

PULSAR: Neon Gravity

A one-thumb neon synthwave reflex game. Tap to flip gravity, thread a glowing orb through hazards, ride an original soundtrack.

iOS App · Swift

Precog

Shipped native iOS app — designed, built, and published end-to-end.

iOS App · Fitness

30-Day Challenger

On-device fitness challenge app with privacy-first design.

Writing & research

Consciousness · framework

TSR

Original interdisciplinary theory, including a Coq formalization.

Available now · advisory, fractional & build — taking a few new engagements this quarter
// Advisory · Fractional · Build · Full-time

Let's build
something.

Tell me what you're trying to ship. Most engagements start with a 30-minute call to see if I'm the right person for it.