About me
Forward deployed AI engineering · LLM systems · agentic AI
I build & deploy production AI systems for broker-dealers.
Forward Deployed AI Engineer — I embed with teams to turn ambiguous business problems into production LLM systems: RAG pipelines, agentic workflows with LangGraph, and multi-agent orchestration that reduce costs, improve decisions, and scale — balancing performance, cost, latency, and compliance.
0
Aircraft fleet supported with ML-driven ops optimisation at IndiGo
0
AI/LLM token cost cut via targeted-surveillance architecture
0
Fewer hallucinations through rigorous AI evaluation practices
0
Person cross-functional team built & led across AI, security & analytics
I’m a Forward Deployed AI Engineer — I build AI systems that production teams actually ship. My work sits where the code meets the customer: embedding with teams, scoping the real problem, and carrying LLM architectures, RAG pipelines, and agentic workflows all the way to production. My background is unusual by design: I came into AI through business and operations, not purely engineering. That shaped how I think — before models, before architecture, before tooling, the real question is always, “what decision are we improving?”
At Frensei (ClawbackVault AI), I designed a three-stage email intelligence pipeline that cut LLM token spend by 80–90% per broker while meeting CASA Tier 2 and Google OAuth Restricted Scope compliance, and built a four-tier signal-detection engine on Claude Sonnet that classifies 15+ behavioural categories with tiered confidence scoring and full audit trails. I partner directly with the CEO on product strategy, pricing architecture, and go-to-market — and built and led a 15-person cross-functional team. Before that, at IndiGo — one of the world’s largest airlines — I led ML-driven optimisation across a 300+ aircraft fleet, delivering ~20% operational cost savings.
I’m completing an MSc in Business Analytics at UCD Michael Smurfit (graduating 2026) while building production AI full-time, with formal AI & ML training from IIT Kanpur. I author open-source Python packages on PyPI — hallucimap, finfeatures, fraudguard — write on AI and strategy, host two podcasts, and am open to Founding AI Engineer, Applied AI Engineer, Forward Deployed Engineer, and AI consulting roles in Ireland, Europe, and globally.
What I do
-
Forward Deployed AI Engineering
Embedding with teams to take AI from zero to production — scoping the business problem, shipping the LLM system, and iterating with real users until it delivers measurable outcomes.
-
AI Systems & LLM Architecture
RAG pipelines, semantic search, and agentic workflows with LangGraph & LangChain — multi-agent orchestration built for performance, cost, and reliability from day one.
-
AI Security & Compliance
AES-256-GCM encryption, PII masking ahead of every LLM call, prompt-injection protection, OAuth 2.0, and GDPR-ready data handling — privacy engineered in, not bolted on.
-
Business Translation & Product Strategy
“What decision are we improving?” — roadmaps, pricing architecture, and go-to-market timing, communicated as fluently to founders as to infrastructure engineers.
Currently building
ClawbackVault AI
Full-Stack · AI · Fintech — protecting broker commissions from silent client churn.
I own the end-to-end build at Frensei: a targeted-surveillance architecture that replaced full-inbox scanning and cut Claude API spend per broker by 80–90%, a three-stage processing pipeline (header scan → PII-masked body fetch → signal classification) where email bodies never touch the database, and a four-tier churn-signal engine powered by Claude Sonnet 4.6 across 15+ behavioural categories.
- Two-pass Gmail / Outlook fetcher via Gmail API & Microsoft Graph (OAuth 2.0)
- “Ghost Audit” — a 60-second inbox scan surfacing commission-at-risk in real time
- AES-256-GCM encryption, PII masking before every LLM call, prompt-injection protection
- Row-level multi-tenant isolation across EU-West & Asia-Pacific; full GDPR export
How ClawbackVault processes a signal
The live pipeline — drag to explore. Email bodies are PII-masked before any LLM call and never touch the database; targeting only the relevant mail is what drives the cost reduction.
Certifications & recognition
-
UCD Advantage Award
University College Dublin · Honor
-
Data Strategy
Professional Certification
-
Google Project Management
Professional Certificate
-
BCG Advanced Analytics & Data Science
Virtual Experience Program
-
Data Visualization & Analytics
Experience Program
-
Relational Data on Azure
Microsoft
-
Google Business Intelligence
Professional Certificate
Tools & stack
AI & LLMs
Data & ML
Platform & cloud
Testimonials