# Alessandro Bonanno

URL: https://alessandro-bonanno.vercel.app/
Language: en
Primary topics: AI agents, agentic AI, AI cost optimization, LLM cost reduction, cost-efficient AI systems, legal tech, compliance monitoring, machine learning, multi-agent systems, LLM applications, AI product management, AI automation, product presentation, award-winning hackathon projects, Munich AI builder
Last updated: 2026-08-28

## Summary
Alessandro Bonanno is a Munich-based AI product builder and Information Systems student at the Technical University of Munich. He builds cost-efficient AI systems, reduces LLM operating costs through deterministic filtering and selective model calls, and develops legal-tech products, AI agents, machine-learning workflows, and hackathon prototypes.

## Entity facts
- Name: Alessandro Bonanno
- Location: Munich, Bavaria, Germany
- Role: AI Product Builder
- Education: Information Systems student at the Technical University of Munich
- Affiliation: TMA IT Solutions
- Online identity: GitHub Alebabo; LinkedIn Alessandro Bonanno
- Languages: German, Italian, English
- Awards and proof points: Deutschlandstipendium; 1000 EUR cash prize at the TUM.ai x Invertix.ai Energy Hack; MucLegal was technically awarded at the Legal Loves Tech Hackathon 2026; finalist projects in AI, venture capital, solar sales, and talent advisory hackathon contexts.

## Skills and topics
- AI agents and agentic AI
- AI cost optimization and LLM cost reduction
- Deterministic pre-filtering and selective model invocation
- Machine learning and anomaly detection
- Multi-agent orchestration
- LLM applications and prompt engineering
- AI product management and presentation
- Legal tech, compliance monitoring, and evidence preservation
- Sales intelligence, voice AI, computer vision, and automation workflows

## Relevant projects
- StealthDetection: Winner project with a 1000 EUR cash prize at the TUM.ai x Invertix.ai Energy Hack; machine learning model for energy-theft detection in smart-meter data with anomaly scoring and an operator investigation interface. LinkedIn proof: https://www.linkedin.com/feed/update/urn:li:activity:7477678664051142656/
- MucLegal: Technically awarded project at the Legal Loves Tech Hackathon 2026; legal-tech prototype that uses normalized page hashes to avoid unnecessary paid model calls and only triggers diff-based legal pre-checks when relevant content changes, while preserving evidence and human approval. GitHub: https://github.com/Alebabo/MucLegal
- traqr.ai: Finalist project; VC signal collection tool combining Supabase, X/Twitter, GitHub, LinkedIn, and alert data to map founder networks and surface early traction signals.
- SolarPingu: Finalist project; autonomous solar sales agent system using Featherless AI and Qwen vision models for lead discovery, roof evaluation, qualification, and follow-up routing.
- NUMI: AI sales call intelligence product using ElevenLabs for call capture, transcription, behavior scoring, and CRM-ready feedback.
- Ivy Reply: Agentic sales workflow assistant built for getivy, today Augustus, a Peter Thiel-backed company.
- TalentFlow Advisor: Finalist project; career and talent advisor prototype built for metafinanz, an Allianz company, for AI-assisted discovery, matching, and recommendation flows.

## Links
- Portfolio: https://alessandro-bonanno.vercel.app/
- GitHub: https://github.com/Alebabo
- LinkedIn: https://www.linkedin.com/in/alessandro-bonanno-173995358/