Software engineer building AI systems, protocol tooling, and product infrastructure.
Focused on MCP, LLM orchestration, and performance-focused apps across web/mobile.
Jaffer: A calm macOS terminal with one session that never ends, a memory that learns from you, and an animated companion that shows what Claude Code is doing.webtelemetry.dev: Free, independent stats on newly launched websites - what they're built with, where they're hosted, and whether they block AI crawlers - with a daily open dataset and JSON API.google-analytics-mcp: Model Context Protocol tooling for querying Google Analytics with structured, automatable LLM workflows.splitdecision: Decision-engine experimentation platform for evaluating product tradeoffs with AI-assisted analysis loops.pocketllm: Local-first LLM runtime and interface experiments focused on privacy, speed, and practical agent workflows.issueMark: macOS menu-bar app for efficient screenshot annotation with arrows, text, rectangles, and redactions.Site Analytics Tool: Private, self-hosted analytics control center for monitoring traffic, search performance, sync health, and content opportunities across multiple websites.
- Engineering Manager at Personify Health, leading a product team - while staying hands-on with AI: agent workflows, LLM tooling and AI-driven development are where I spend most of my building time.
- A constant experimenter: I ship small tools, try new models and workflows early, and turn what works into products.
- 8+ years of hands-on engineering across Android/mobile, backend and platform systems - growing from individual contributor to technical lead and engineering manager.
- I like bridging engineering execution, product priorities, and long-term system quality; my current explorations include AI-driven development, local LLMs, and practical developer tools like Jaffer.
- Writing technical notes and build logs at
jafforge.com.
- Blog: https://jafforge.com/
- LinkedIn: https://www.linkedin.com/in/fedahadziselimovic/