Skip to content
STATUS: open to new projects

$ ./introduce --self

Fredy Rodriguez

>

I build intelligent systems that actually ship — LLMs running on bare metal I manage myself, cloud infrastructure squeezed for every megabyte, and data pipelines that hold up in production. Broadcasting from Prague, CZ.

Explore my builds LinkedIn GitHub
fredy@ingfrerod:~$ systemctl status
●

ingfrerod.web.service active (running)

Hugo static · NGINX · OCI x86 · 1 vCPU / 1 GB

●

llm.inference.service active (running)

Gemma 4 26B GGUF · llama.cpp · Ampere ARM 4c / 24 GB

●

apps.streamlit.service active (running)

api · app · trading · TLS via Let's Encrypt

0B

params local

0$

cloud bill

0%

self-hosted

fredy@ingfrerod:~$

scroll
Python FastAPI Streamlit llama.cpp GGUF PyTorch Pandas LangChain Docker NGINX Linux Oracle Cloud Terraform PostgreSQL Redis Git CI/CD Raspberry Pi MQTT IoT Arduino REST APIs Hugo Tailwind Python FastAPI Streamlit llama.cpp GGUF PyTorch Pandas LangChain Docker NGINX Linux Oracle Cloud Terraform PostgreSQL Redis Git CI/CD Raspberry Pi MQTT IoT Arduino REST APIs Hugo Tailwind
01 / About me

Curriculum Vitae

A look into my professional background, skills, and experience — from building smart systems with Python to cloud-native deployments and AI integrations.

Python Cloud & DevOps AI Systems Data Engineering Robotics & IoT

I specialize in creating efficient, scalable solutions for modern tech challenges. My work spans Python backends, cloud-native deployments, and AI integrations — with a strong bias toward systems that run lean. The proof is this site: it and a 26-billion-parameter language model both live on free-tier hardware I provision, harden, and maintain myself.

Python & Backend95%
Cloud & Infrastructure90%
AI / LLM Systems88%
Data Engineering85%
View full profile on LinkedIn

Location

Prague, Czech Republic

Work status

Open to new projects

🌍 Languages

  • Spanishnative
  • Englishfluent
  • Portuguesefluent
  • Czechcommunicative

💡 Fun fact

☕ I love coffee, and usually I drink my own coffee from my family plantation in Colombia. Nothing beats starting the day with a perfect cup from home.

🎯 Hobbies & interests

🎾
Tennis
🏸
Badminton
🔧
Electronics & IoT
🤖
AI Research
02 / What I've built

Featured Projects

Live systems, not screenshots. Every project below runs right now on infrastructure I operate.

💬 live

AI Chat API

Securely chat with advanced language models through a simple login system. Choose from DeepSeek-V3, DeepSeek-R1 (Reasoner), and Gemini 2.0 Flash — each tailored for different AI tasks.

Python Streamlit LLM APIs JWT Auth
View project
🦙 live

Locally-Hosted LLM Chat

Private chat app running entirely on my own server, powered by Google Gemma 4 26B in GGUF format. Fully offline inference on an Oracle Cloud Ampere ARM instance (4 cores, 24 GB RAM), served securely behind NGINX with HTTPS.

llama.cpp Gemma 4 26B Ampere ARM NGINX + TLS
Try the demo
📈 live

Trading 212 Portfolio Tracker

Comprehensive dashboard for tracking and analyzing Trading 212 investments. Portfolio breakdowns, ETF vs. stock allocation, dividend history tracking, and performance comparisons.

Python Streamlit Data Viz Trading212 API
View dashboard
03 / Under the hood

The Machines Behind It

No managed platforms, no vendor magic. Two free-tier instances, provisioned and tuned by hand.

Node 01 · x86

Web & edge

cpu
1 vCPU AMD
memory
1 GB
stack
Hugo + NGINX
tls
Let's Encrypt · auto-renew

Migrated off WordPress to a static Hugo build — dropped memory use from ~689 MB to a rounding error, and page loads to a single gzipped request.

Node 02 · ARM64

Inference

cpu
4 cores Ampere Altra
memory
24 GB
runtime
llama.cpp · ARM NEON
model
Gemma 4 26B (GGUF)

A 26-billion-parameter model doing CPU-only inference, with zero GPU and zero external API calls. Every token is generated on hardware I control.

$ open --connection

Let's build something intelligent

Available for Python, cloud architecture, and AI systems work — remote from Prague, or on-site across the EU.

Send me a message Connect on LinkedIn