Medium

Ida Silfverskiöld

Pinned

In

Level Up Coding

by

Ida Silfverskiöld

·

Sep 30, 2025

Agentic AI: Implementing Long-Term Memory

This article helps you understand how to build complex memory systems and the frameworks that help

Agentic AI: Implementing Long-Term Memory

Pinned

In

Data Science Collective

by

Ida Silfverskiöld

·

Apr 15, 2025

Agentic AI: Comparing New Open-Source Frameworks

By looking at functionality and learning curves

Agentic AI: Comparing New Open-Source Frameworks

Pinned

In

TDS Archive

by

Ida Silfverskiöld

·

Dec 23, 2024

Advanced Prompt Engineering: Chain of Thought (CoT)

Comparing different techniques for reasoning

Advanced Prompt Engineering: Chain of Thought (CoT)

Pinned

In

TDS Archive

by

Ida Silfverskiöld

·

Nov 12, 2024

Economics of Hosting Open Source LLMs

Leveraging various deployment options

Economics of Hosting Open Source LLMs

Pinned

In

TDS Archive

by

Ida Silfverskiöld

·

May 28, 2024

Fine-Tune Smaller Transformer Models: Text Classification

Using Microsoft’s Phi-3 to generate synthetic data

Fine-Tune Smaller Transformer Models: Text Classification

In

Data Science Collective

by

Ida Silfverskiöld

·

6d ago

Building the Harness Around AI Coding Tools

The basics like permissions & sandboxing, AGENTS.md, rules, docs, memory, planning, subagents & skills

Building the Harness Around AI Coding Tools

In

Level Up Coding

by

Ida Silfverskiöld

·

Sep 13

On Building an Open Source AI Video-to-Audio Editor

Something that fixes the whole voice drift thing

On Building an Open Source AI Video-to-Audio Editor

In

Level Up Coding

by

Ida Silfverskiöld

·

Jul 12

I Spent the Summer Testing 14 OCR Engines

To figure out where each excelled and where it failed

I Spent the Summer Testing 14 OCR Engines

In

Data Science Collective

by

Ida Silfverskiöld

·

May 5

Agentic AI: How to Save on Tokens

Caching, lazy-loading, routing, compaction, and so on

Agentic AI: How to Save on Tokens

In

Data Science Collective

by

Ida Silfverskiöld

·

Feb 17

Testing a Naive RAG Pipeline vs an ‘Advanced’ One

Which wins where and why?

Testing a Naive RAG Pipeline vs an ‘Advanced’ One

Ida Silfverskiöld

Ida Silfverskiöld

Friend of Medium

Building stuff (applied AI & general tech). Connect 🤝 linkedin.com/in/ida-silfverskiold, github.com/ilsilfverskiold, ilsilfverskiold.com

Following

Help

Status

About

Careers

Press

Blog

Store

Privacy

Rules

Terms

Text to speech