Chris France

Persistent AI memory with vector similarity retrieval. Memories reinforce with use, decay over time, and resurface automatically when relevant.

Automated LLM vulnerability testing across 10 attack categories — prompt injection, jailbreaking, data extraction. Scans local and cloud models.

Private document-grounded chat. Upload documents, ask questions, get sourced answers. Fully local — no API keys, no cloud dependencies.

Total cost of ownership for GPU builds vs cloud inference. Break-even analysis across 6 hardware tiers over 36 months.

Monte Carlo capacity forecasting across compute, storage, power, and cooling. AI-powered scenario analysis with decision timelines.

Datacenter investment analysis for PE/VC. Financial modeling, M&A valuation, and deal memos across 14 global markets.

Transit asset management platform built on public MBTA data. FTA condition scoring, capital planning, and state-of-good-repair analysis across 200+ assets.

12-lecture executive education course. Strategy frameworks, live demos, and implementation playbooks for enterprise AI adoption.

Full-stack model health scoring with 31 automated metrics, 209 passing tests, C# cloud extraction pipeline, and 15 dashboard pages.

Winning hackathon entry at Autodesk DevCon 2026 (Amsterdam). AI-powered MEP audit tool that detects dead legs in hospital domestic-water systems — a leading source of Legionnaires' disease. Extracts Revit topology via Dynamo, sends it to a remote MCP server for Claude analysis, then colors elements in the model by risk severity. "AI saves lives" theme — humanitarian framing won the room.

Adversarial security suite that thinks like an attacker, not a developer. 217 individual tests across 18 attack categories — JWT forgery, RBAC and IDOR escalation, SQL/code/path injection, file-upload abuse, race conditions, supply-chain compromise, Linux privesc, post-exploitation, tunneling bypass, LLM jailbreaks, container escapes, TLS & header hardening, and AI-powered source-code review via Claude Code Security. Every production app I ship is pushed through this suite as a build gate before it goes live — the count keeps growing as new attack patterns surface.

Infrastructure control plane for AI-powered apps. App registry, GPU monitoring, deploy/rollback, LLM load balancing, API cost tracking, and an embedded AI ops assistant — all from one dashboard.

Enterprise lifecycle platform tracking projects from BD pipeline through proposals, execution, and closeout. Role-gated dashboards, win/loss analytics, and lessons learned engine.

IP-protected email triage across multiple accounts with local LLM classification, calendar-aware meeting prep, real-time action item extraction from transcripts, and prioritized briefings — all processed on-device.