AI Engineer · Agentic AI & LLM Applications

Branimir Antičević

AI Engineer who designed, built, and commercially shipped a production-grade AI product with real users.

  • Commercially launched — real paying users in production workflows.
  • 35+ schema-validated tools
  • End-to-end delivery with security & licensing
View ProjectsOpen Career Assistant →
Got recruiter questions? Ask my AI assistant→

Proof of work

Real systems, real implementations, real users.

Real-time AI system integrated into Ableton Live, used in production workflows.

Commercially launched · v2.0.1Agent Systems

Profesor Abelton — AI Copilot for Ableton Live

A standalone desktop AI that turns natural language into Ableton Live actions — built to feel like a friend who actually knows Ableton.

Shipped to paying users in less than 3 months from first line of code — music producers running it inside their day-to-day Ableton Live sessions.

  • Streaming full Ableton session state (tracks, clips, devices, tempo) in real time without blocking the UI.
  • Designing 35 schema-validated MCP tools so the LLM can act safely and predictably.
  • Hardening the command pipeline: a 40-action allowlist, parameter sanitization, and a 12-command batch limit.
  • Making it a real product: Fernet-encrypted API keys, machine-bound Gumroad licensing, and a 5-step first-launch wizard.
PythonMCPClaudeGroqWebSocketsFernetPyInstaller
Gumroad ↗GitHub ↗
v2.0.1
Commercially launched
35
Schema-validated MCP tools
2 LLMs
Claude + Groq routing
Passed
Structured security audit
Win + macOS
Distributed via Gumroad
0
Bandit medium/high findings

Static analysis via Bandit (0 medium/high findings) + structured security audit. Supplementary LLM-assisted review checks were also run as part of the audit methodology (GPT-4o: 88/100, Grok: 80/100).

Production systems, not demos.

Ready to build something similar?

I help companies build AI agents, workflow automations, API integrations, and production-grade AI applications. Whether you need an AI copilot, internal automation system, intelligent workflow orchestration, or a custom LLM-powered tool, I can help design and implement it.

Available for

  • AI Automation Engineering
  • Agent Systems
  • API Integrations
  • LLM Applications
  • Technical Consulting
Contact meView GitHub ↗

What I build

Services & technical capabilities

The kinds of systems I design and ship — end to end.

AI Agents

Agents that read live state, reason over it, and take actions through validated tools.

Agentic Workflows

Multi-step LLM workflows with tool-calling, guardrails, and deterministic fallbacks.

MCP Integrations

Custom Model Context Protocol servers exposing schema-validated tools to LLMs.

RAG Systems

Retrieval-augmented assistants grounded strictly in your data — like the one on this site.

API Integrations

Real-time integrations with local and third-party APIs over REST and WebSockets.

Full-Stack AI Apps

Next.js + TypeScript front ends backed by Python AI services.

Python Backends

Async Python services with FastAPI, WebSockets, and careful error handling.

Workflow Automation

Python automation for trading, on-chain workflows, and data pipelines.

LLM-Powered Tools

Desktop and web tools built on Claude, Groq, OpenAI, and xAI Grok.

Why companies hire me

Built to ship, not just to demo.

I ship production AI systems end-to-end — not just prototypes. Here's what that actually looks like in practice.

One commercially launched AI product in production

Designed, built, and shipped Profesor Abelton — a desktop AI copilot for Ableton Live, commercially launched with real paying users in production workflows. Includes 35+ schema-validated tools, real-time session streaming, encryption, command allowlists, and commercial licensing.

Production-grade, not demo-grade

Security reviews, encrypted API keys, machine-bound licensing, parameter sanitization, and strict command guardrails were part of the first release — not added later.

Ships from vague to working

Comfortable taking a rough idea or one-line brief and turning it into a usable product with real users. Solo end-to-end delivery (architecture, implementation, security, distribution).

Understands both sides of the table

16 years in journalism and marketing before switching to AI. I build for business outcomes, not just technical elegance, and can communicate clearly with both engineers and non-technical stakeholders.

Communicates like an engineer, thinks like a product person

16 years writing for audiences means clear async communication, precise documentation, and the ability to translate technical decisions for non-technical stakeholders. No hand-holding required.

Remote-first and independent

Native Croatian, professional English. Experienced working solo with full ownership of a commercial product.

From newsroom to shipped AI product— read the full story →

Stack & skills

Technologies I work with

AI & LLM Systems

  • Model Context Protocol (MCP)
  • Tool / Function Calling
  • Agentic Workflows
  • RAG & Embeddings
  • Claude API
  • Groq API

Languages

  • Python
  • C#
  • JavaScript / TypeScript
  • Node.js
  • Rust

Software Engineering

  • Async Python
  • WebSockets
  • REST APIs
  • FastAPI
  • Real-Time Integrations
  • Next.js

Security & Production

  • Fernet Encryption
  • Command Allowlisting
  • Parameter Validation
  • License Systems
  • Static Analysis (Bandit)
  • Secure Local Deployments

Ask anything about my work.

A grounded assistant answers recruiter questions and evaluates role fit — based strictly on my real experience.

Open Career Assistant →