New in the Claude Marketplace: @augmentcode, @boltdotnew, @coderabbitai, @hebbia, and @WeAreLegora.
Apply your existing Anthropic spend commitment toward their Claude-powered products.
Learn more: https://t.co/J3Sdsg2lU6
Built my own private cloud with Docker 🐳
One YAML file deploys: → Nextcloud (Google Drive replacement) → Vaultwarden (password manager) → Jellyfin (media server) → Uptime Kuma (monitoring)
Zero subscriptions. Full control.
https://t.co/S6PWyLpqoZ
Just shipped my CVD Risk Predictor, an ML-powered cardiovascular screening tool built from my MSc dissertation!
308K patient records Gradient Boosting + SMOTE (AUC: 0.827) Tiered treatment recommendations (T1–T5) React + FastAPI + deployed on Vercel.
https://t.co/JxhEKDFLVS
Built an AI Resume Analyzer.
Upload CV + paste job description → get a match score, skill gap analysis, and specific fixes.
Built it because I needed it myself. Now it's live for everyone.
Python + Groq API. Link in bio.
#AI#BuildInPublic#OpenToWork
My portfolio site is live.
I built it to have one place where all my AI/ML work lives — projects, skills, and what I'm building next.
What's on there right now:
→ Brain Image Classification — ensemble deep learning system for tumor detection from MRI scans
More projects coming. Currently finishing my MSc in Artificial Intelligence at the National College of Ireland.
Check it out: https://t.co/vRd6ykLBxd
GitHub: https://t.co/hc4yYZEsYU
Built a RAG-powered research assistant using Cohere's API.
Upload docs → ask questions in plain English → get answers with citations.
Cohere Embed + ChromaDB + Command model. Full pipeline, deployed live.
Try it: https://t.co/voQ8vnEWfw
@cohere @coaborhere.
#AI#RAG#Cohere
Built a neural network from scratch with just NumPy.
No PyTorch. No TensorFlow. Just math and code.
Forward pass → loss → backprop → gradient descent. All manual.
If you can't explain it from scratch, you don't really understand it.
Code: https://t.co/r9eQwh1mLD
#AI#ML#DL
Built a multi-agent AI data analyst 🤖
Upload any CSV + ask a question in English → 4 agents collaborate:
Profiler analyzes quality
Cleaner fixes issues
Analyst writes & runs real pandas code
The reporter writes a summary
Github link: https://t.co/OKfT12eze7
The Analyst agent doesn't guess — it writes actual Python code, executes it with pandas/matplotlib, and returns computed results + charts.
Ask "highest revenue product?" → it runs df.groupby().sum() → returns $97k for Laptops.
That's the difference between a wrapper and a tool.
New Anthropic research: Project Deal.
We created a marketplace for employees in our San Francisco office, with one big twist. We tasked Claude with buying, selling and negotiating on our colleagues’ behalf.
Completed the AWS course: Fundamentals of Machine Learning & Artificial Intelligence.
Strengthened my understanding of core ML concepts, model types, and AI fundamentals.
Continuing to build deeper in this space 🚀
Completed “Understanding ChatGPT” on DataCamp.
Learned how LLMs work, prompt design basics, and real-world use cases.
Exploring more in AI and generative models next.
Completed AWS ML Engineer Associate Curriculum Overview.
Gained clarity on the ML engineer role, workflows, and AWS-based ML pipelines.
Moving closer toward hands-on implementation.