Towards AI

“Towards AI” is a platform that delves into various aspects of artificial intelligence and machine learning. The content covers topics such as data augmentation for machine learning, Python programming scripts, and the use of Python in web3, blockchain, and smart contracts. Additionally, it explores the challenges and solutions in building AI applications with features like short-term memory and persistent conversation history. The platform also discusses the importance of understanding and utilizing enterprise-grade tools like Spark, EMR on EKS, and Airflow 3 for creating secure and efficient knowledge bases for AI applications.

Visit Towards AI

Build a Jev-Style Decision Model from an LLM, Part 4: Letting the Backbone Learn, Then Giving It…

Towards AI

Build a Jev-Style Decision Model from an LLM, Part 4: Letting the Backbone Learn, Then Giving It Eyes Letting the backbone learn with LoRA and then giving the same scoring head images to read. Photo ...

Build a Jev-Style Decision Model from an LLM, Part 3: Training the Jev Head

Towards AI

How four vectors become probabilities: the pointer head that scores the options, how it is trained and how far a frozen backbone can take it. Photo by Alex Knight on Unsplash Where Part 2 left off Pa...

Why Google’s Small EmbeddingGemma 2 Model Deserves More Attention

Towards AI

EmbeddingGemma 2 packs text, code, image, video and audio search into 270M to 740M parameters. See what it does well and where it falls… Continue reading on Towards AI

Build a Jev-Style Decision Model from an LLM, Part 2: Opening the Model

Towards AI

What is inside a hybrid vision-language model, what that rules out and how to feed it a decision. Photo by Tim Mossholder on Unsplash Where Part 1 left off Build a Jev-Style Decision Model from an LL...

Copilot’s Fabric Token Rates Dropped. On October 1, Consumption Changes Again.

Towards AI

As of 18 September 2026, Microsoft’s public Copilot-in-Fabric documentation lists 100 CU seconds per 1,000 input tokens, 10 CU seconds per… Continue reading on Towards AI

I Built a Code Search Engine With EmbeddingGemma 2 in 80 Lines of Python

Towards AI

Semantic code search that runs locally. Chunk, embed, and search your repo by meaning with EmbeddingGemma 2. Continue reading on Towards AI

Sophi Harness — Calling in a subcontractor: subagents as just another Tool

Towards AI

Series: Building SoPhi — a Kotlin Agent Harness — Article 08 A general contractor who needs an electrician doesn’t invent a new business process to get one. They make a phone call, describe the job, ...

The KV Cache Is Just Session State: Disaggregated Inference for Backend Engineers

Towards AI

Prefill, decode, and why routing an LLM request is closer to routing a shopping cart than a stateless API call Disclosure: I drafted parts of this article with AI writing tools. I edited all of it, c...

Jakob’s Law in the Age of AI Chat: 5 Rules for Designers

Towards AI

Your users already learned how to use AI from a chat box, so here is how to design for that habit without trapping them in it. Continue reading on Towards AI

Gaussian Splatting for Self-Driving Simulation, Built in Plain PyTorch on a Laptop GPU

Towards AI

Author: Tamoghna Das Featured image: Left: a camera photo not used in training. Right: the same view, rendered from the trained 3D Gaussian model. License plates blurred. Source: Image by the author, ...

17 Days with Jev: I Put a Sub-Second “System One” AI on a Real Trading Account

Towards AI

Jev doesn’t write text. It reads a state, answers typed questions and hands back probabilities in a third of a second. Here’s how it works, how to build with it, and what happened when it traded real...

I Ran 3 Office Tasks on 3 AI Agents, 27 Times. Here’s What It Cost

Towards AI

Lindy, Manus and Claude Free on the same contract summary, CSV reconciliation and cold email sequence: median cost per task, pass rates… Continue reading on Towards AI