Engineering leader and researcher (PhD, Computer Software Engineering) working at the intersection of applied AI, machine learning, and mathematical modeling, with a production background in backend systems, SDKs, developer platforms, distributed systems, and blockchain infrastructure.

I design and ship production systems in Go, Rust, Python, and TypeScript, and my research sits where mathematics meets machine learning:

• deep reinforcement learning and autonomous decision systems
• LLM agents: orchestration, evaluation, and benchmark design
• applied mathematics: optimization, stochastic modeling, and simulation
• market mechanisms and blockchain infrastructure

PhD in Computer Software Engineering from the Technological University of the Shannon (2026), on AI and autonomous decision systems. My research spans deep reinforcement learning for sequential decision problems, LLM orchestration with deterministic safety guarantees, mathematical modeling of market mechanisms (automated market makers, concentrated liquidity), and the evaluation of LLM agents on scientific and mathematical tasks.

Researcher with the Montana Research Foundation since 2023. Open-source contributor to Terminal-Bench; benchmark contribution accepted into Terminal-Bench Science, the scientific-domain extension of the benchmark.

• Bounded LLM Orchestration with Deterministic Safety Gates for Concentrated Liquidity Management — IEEE Access, 2026
• From MDP to POMDP for Concentrated Liquidity Management: A Synthetic-to-Historical Deep Reinforcement Learning Strategy — IEEE International Conference on Consumer Electronics (ICCE), 2026
• Optimizing Concentrated Liquidity Management: A Synthetic-to-Historical Deep Reinforcement Learning Strategy — IEEE International Conference on Decentralized Applications (DAPPS), 2025
• CARM: A Blockchain-based Content Quality Assessment and Rewarding Mechanism — IEEE International Conference on Blockchain, 2021

• Research in deep reinforcement learning, LLM orchestration, and agent evaluation
• Engineering leadership across AI initiatives, AI workflows, backend systems, and platform services
• Evaluation-driven development for agentic systems; Claude Certified (Anthropic)

Engineering leader with 10+ years across software engineering, machine learning, platform architecture, and Web3: AI initiatives and workflows, backend and platform services, SDKs, client systems, developer platforms, chain infrastructure, and AI tooling, in both product environments and consulting engagements, including architecture ownership for early-stage and revenue-generating platforms.

• Claude Certified (Anthropic)
• Associated Professional Member (APM), BCS, The Chartered Institute for IT

• Languages: Go, Rust, Python, TypeScript
• AI / ML: deep reinforcement learning, agent orchestration, agent evaluation and benchmark design, Claude, OpenAI, Llama, Ollama
• Systems: event-driven backends, REST APIs, PostgreSQL, MongoDB, DynamoDB, CI/CD (GitHub Actions), React / Next.js
• Web3: Solidity, Hardhat, Ethers.js, IPFS, Ethereum, Cronos, subgraphs

• GitHub: @rarcifa
• LinkedIn: ricardogiuliano