Former Reuters/Bloomberg engineer pivoting to AI/ML engineering in buy-side finance.
I build production-grade RAG systems, intelligent automation, and data infrastructure that solve real workflow problems for portfolio managers, analysts, and data teams at asset managers and private banks.
RAG Systems • Intelligent Automation • Financial Data Infrastructure
My focus: bridging financial domain expertise with modern AI capabilities to create tools that actually get adopted and drive measurable outcomes.
Problem: Sales teams and PMs lose 5-8 hours weekly digitizing meeting notes into CRM systems
Solution: End-to-end RAG pipeline with speaker identification, metadata extraction, and conversational search via Telegram
Impact: 75% time reduction, 250+ meetings processed, 100% automated CRM data capture
Tech: Python • LangChain • PostgreSQL+pgvector • Gemini Pro • Claude • n8n
📄 Full Case Study | 🔗 Live Demo (upon request)
Problem: Portfolio managers spend 15-20 hours weekly on routine research across 20-100 positions
Solution: AI-powered scheduled workflows generating daily market intelligence, risk alerts, and research memos with confidence scoring
Impact: 60-70% time savings, 12-24 hour information lead time on material portfolio events, scalable to 100+ positions
Tech: Python • GPT-4 • Claude • PostgreSQL • n8n • Jupyter • Docker • Webhooks
📄 Full Case Study | 🔗 Live Demo (upon request)
Problem: Organizations face 10x variance in AI output quality due to poor prompting skills
Solution: Real-time prompt analysis system with multi-dimensional scoring, thematic categorization, and self-serve analytics dashboard
Impact: 20% quality improvement over 3 months, 40% reduction in wasted iterations
Tech: Python • Gemini 2.0 • PostgreSQL • Metabase • n8n • Docker
📄 Full Case Study | 🔗 Live Demo (upon request)
Current: Building AI systems for financial workflows (personal projects + consulting)
Previously:
- Reuters: Market data infrastructure, real-time systems
- Bloomberg: Terminal features, quatitative modelling, data pipeline engineering
- swissQuant: Portfolio analytics & Optimization platforms
Domain Expertise: Market data • Portfolio management workflows • Investment research • Risk analysis • Trading systems
AI & LLM:
OpenAI GPT-4 • Anthropic Claude • Google Gemini • LangChain • RAG Systems • Prompt Engineering
Data Infrastructure:
PostgreSQL • Vector Databases (pgvector) • SQL • Data Modeling • ETL Pipelines
Orchestration & Automation:
n8n • Webhook APIs • Event-Driven Architecture • Scheduled Workflows
Development & Deployment:
Python • Docker • Git • Self-Hosted Infrastructure • API Integration • CI/CD
Financial Domain:
Market Data (Reuters/Bloomberg) • Portfolio Management • Investment Research • Risk Analysis • CRM Systems (Salesforce)
Analytics & Visualization:
Jupyter Notebooks • Metabase • Dashboard Design • Data Storytelling
1. Domain-Driven Design
Every system is built around actual financial workflows, not generic AI demos. I understand PM/analyst pain points because I've worked in financial data infrastructure for years.
2. Production Quality from Day One
My projects include error handling, monitoring, audit trails, and cost optimization. They're designed to be deployed, not just prototyped.
3. Measurable Business Impact
All projects have clear KPIs: time savings, accuracy improvements, adoption metrics. I build to solve problems, not to showcase technology.
4. Cost-Conscious Architecture
Self-hosted where it makes sense, leveraging open-source tools. My Portfolio Intelligence System costs $10/month vs $500+ for SaaS alternatives.
- Advanced RAG architectures (hybrid search, reranking, query decomposition)
- Agentic workflows with n8n and Claude/Gemini (prompt chaining, orchestration-worker, agents)
- Production ML observability and evaluation frameworks
- Financial time series analysis with transformer models
Open to opportunities in:
- Hedge Funds & Asset Managers (AI/ML Engineering, Quantitative Development)
- Private Banks (Data Infrastructure, Analytics Teams)
- Fintech (AI-powered research tools, data platforms)
Ideal roles:
AI/ML Solutions Engineer • Quantitative Strategist • Data Infrastructure Engineer
🔗 Links:
LinkedIn • Portfolio PDF
💡 Why I'm pivoting to buy-side AI:
I spent years building data infrastructure at market data vendors. I know the tooling, the workflows, the pain points. Now I'm applying modern AI capabilities to solve the problems I watched portfolio managers and analysts struggle with daily.
The future belongs to firms that augment human judgment with intelligent systems—not replace it, but systematically enhance it. That's what I build.
Last updated: December 2025