Iman Emrani

AI and LLM engineer. I build LLM applications that run in production, and I work directly with the people who use them.

Authorised to work in Italy. Available worldwide as a remote contractor.

Selected work

Global Campaign Localisation Copilot

An LLM application for multi-market marketing operations. It extracts a structured campaign spec from a brief and a set of brand guidelines, generates channel- and market-specific localised copy, and runs an automated brand-compliance check with a human approval queue and a full audit trail. Deployed and running on Railway with Postgres. It ships with an interactive masterclass, in English and Italian, that explains the architecture and the governance decisions to a non-specialist audience.

Python, FastAPI, SQLAlchemy, Anthropic SDK, Pydantic, React, Postgres, Railway

Overview and live app · Source

Also built

arbscopeArbitrage-opportunity detection across crypto venues, modelling cross-venue spreads with inter-leg timing and execution costs to determine whether a detected opportunity is actually capturable.
BitfaSignalA trading-signal research and backtesting harness: rule discovery over OHLCV data with out-of-sample validation and fee-aware evaluation, reporting negative results rather than over-fitting them into apparent performance.
GenovaAn automated multi-source content aggregation and digest pipeline: scraping, deduplication, change detection, LLM summarisation and scheduled delivery, built to stay correct under partial failure.