💡 I build AI systems for high-stakes environments—engineering reasoning, risk-aware, and autonomous architectures for finance, decision intelligence, and large-scale institutional systems.
I specialize in transforming advanced mathematical, probabilistic, and philosophical concepts into production-grade AI infrastructure that can reason under uncertainty, manage risk, and operate reliably at scale.
As a Senior AI Systems Engineer, I design and deploy intelligent architectures for domains where reasoning quality, uncertainty management, and reliability are critical.
My work focuses on:
- Generative AI and LLM reasoning systems
- Retrieval-augmented and agentic architectures
- Financial intelligence and tail-risk modeling
- Autonomous decision systems
- AI infrastructure and scalable inference pipelines
I work at the intersection of:
- AI engineering
- probabilistic and quantitative reasoning
- systems architecture
- quantitative finance
- institutional-scale intelligence systems
I do not build AI demos. I build systems designed to operate under real-world ambiguity, adversarial conditions, and complex decision environments.
Generative AI & Reasoning Systems — RAG Architectures · Structured Reasoning · Multi-Agent Systems · LangGraph · RLHF · Context Engineering · LLM Evaluation
AI & Machine Learning — Deep Learning · Reinforcement Learning · Anomaly Detection · Optimization Algorithms · Probabilistic Modeling
Financial Intelligence & Quantitative Systems — Tail-Risk Analysis · Portfolio Optimization · Algorithmic Trading · Risk Modeling · Market Simulation
Infrastructure & Scalable AI Systems — Distributed Systems · Scalable Inference · Vector Databases · Qdrant · GPU Workloads · Observability · MLOps · LLMOps
Data & Backend Engineering — Python · FastAPI · Pandas · MongoDB · Docker · CUDA · PyTorch · Transformers
Problem Solving at Scale — I decompose complex, high-dimensional problems into tractable components and build AI systems that handle real-world scale, noise, and ambiguity.
Risk Analysis — I think in probabilities and downside scenarios, with a focus on fat-tail modeling. I design systems that don't just optimize for expected reward, but deeply understand and manage extreme events where conventional models fail.
Systematic Thinking — I approach problems with structured, methodical frameworks—mapping cause and effect, identifying dependencies, and building reproducible pipelines from chaos.
Nonlinear Thinking — I identify hidden dependencies, second-order effects, and unconventional solution paths across AI, finance, infrastructure, and institutional systems. I focus on architectures that remain robust under uncertainty, scale, and adversarial conditions.
A language model–based reasoning engine that processes complex scenarios, evaluates multi-step logical chains, and produces structured, defensible conclusions. Built to make LLMs reason through ambiguity—not just generate text.
An implementation of the RAFT (Retrieval Augmented Fine-Tuning) paradigm for LLMs, combining domain-specific supervised fine-tuning with in-context retrieval. RAFT-LM trains models to intelligently distinguish between relevant and distractor documents, significantly improving answer accuracy in "open-book" domain-specific applications.
High-performance tools for extreme event analysis in financial and statistical domains. Focused on the tail of the distribution—where risk lives, conventional models fail, and precise AI methods are required.
An adaptive data augmentation framework designed to dynamically select and apply augmentation strategies based on dataset characteristics and model feedback. RADA moves beyond static, one-size-fits-all augmentations, using a contextual approach to optimize the training pipeline, improve model generalization, and enhance robustness across diverse domains.
- Ph.D. — Artificial Intelligence
- M.Sc. — Computer Engineering (AI)
- B.Sc. — Software Engineering
Exploring autonomous financial intelligence systems that combine:
- explainable AI
- probabilistic reasoning
- agentic workflows
- risk-aware architectures
- institutional-scale decision systems
Researching AI-mediated governance and large-scale coordination systems inspired by:
- Karl Popper’s Open Society
- computational governance
- adaptive institutional design
- algorithmic social contracts
- decentralized intelligence systems
Relevant concepts and adjacent research:
- Society-in-the-Loop (SITL)
- Constitutional AI
- AI governance frameworks
- Open institutional architectures
I believe the next generation of AI systems will not simply generate content—they will coordinate institutions, manage uncertainty, reason under incomplete information, and augment large-scale decision systems.
My focus is building AI architectures that are:
- reliable under uncertainty
- explainable in high-stakes environments
- robust against adversarial conditions
- scalable across institutional contexts
- grounded in systems thinking rather than hype
I am particularly interested in the intersection of:
- AI reasoning
- governance systems
- financial intelligence
- institutional architecture
- computational social systems
Thanks for stopping by. Let's build the future of AI together. 🚀