The Curious Engineer

All blogs / AI / ML

philschmid.de

philschmid.de

Personal Blog of Philipp Schmid Staff Engineer at Google DeepMind.

AI / MLPythonDev tools / editors / IDEs

’20’21’22’23’24’25’26
Posts
214 from 26 Feb 2020 to 28 Sep 2026
Posting
weekly (41 in the last year)
Kind
personal blog
Feed
https://www.philschmid.de/rss

2026 35

  1. Credentials API for Gemini Managed Agents
  2. Create your own voice for Gemini 3.8 TTS
  3. How Foundational Models Became Superhuman in Bash
  4. Setup OpenClaw 2.0 with Gemini in Under 60 Seconds
  5. Recursive Self-Improvement
  6. Controlling Android with Gemini 3.7 Flash and 150 lines of Python
  7. Google Maps and Google Search now work together in the Gemini API
  8. Evaluating Agents Beyond the First Prompt
  9. Building Managed Agents That Use GitHub Without Exposing Your Token
  10. Control an Android Phone with Gemini 3.5 Flash Computer Use
  11. Getting started with the Gemini Interactions API
  12. How Gemini Managed Agents Works under the Hood
  13. Gemini Managed Agents: Developer Guide
  14. How Agents Manage Other Agents: Four Subagents Patterns in 2026
  15. How to use Deep Research with the Gemini API
  16. How to correctly use MCP servers with your AI Agents
  17. 8 Tips for Writing Agent Skills
  18. How to use Gemma 4 with the Gemini API and Google AI Studio
  19. How Kimi, Cursor, and Chroma Train Agentic Models with RL
  20. Combine Built-in Tools and Function Calling in the Gemini Interactions API
  21. Developer Guide: Nano Banana 2 with the Gemini Interactions API
  22. How Autoresearch will change Small Language Models adoption
  23. Practical Guide to Evaluating and Testing Agent Skills
  24. Writing a Good AGENTS.md
  25. Agents: Inner Loop vs Outer Loop
  26. Can We Close the Loop in 2026?
  27. Multimodal Function Calling with Gemini 3 and Interactions API
  28. Getting Started with Gemini Deep Research API
  29. The Agent Client Protocol Overview
  30. Gemini Interactions API Quick Start
  31. MCP is Not the Problem, It's your Server: Best Practices for Building MCP Servers
  32. Transparent PNG Stickers with Nano Banana Pro and Gemini interactions API
  33. Building Agents with the Gemini Interactions API
  34. Introducing MCP CLI: A way to call MCP Servers Efficiently
  35. The importance of Agent Harness in 2026

2025 30

  1. 8 Predictions for 2026. What comes next in AI?
  2. Context Engineering for AI Agents: Part 2
  3. Why (Senior) Engineers Struggle to Build AI Agents
  4. Practical Guide on how to build an Agent from scratch with Gemini 3
  5. Gemini 3 Prompting: Best Practices for General Usage
  6. Gemini API File Search: A Web Developer Tutorial
  7. Build your first AI Agent with Gemini, n8n and Google Cloud Run
  8. AI Agent Benchmark Compendium
  9. Agents 2.0: From Shallow Loops to Deep Agents
  10. The Rise of Subagents
  11. The 10 Steps for product AI generation with Gemini 2.5 Flash
  12. Memory in Agents, Make LLMs remember.
  13. Google Gemini CLI Cheatsheet
  14. Code Sandbox MCP: A Simple Code Interpreter for Your AI Agents
  15. Integrating Long-Term Memory with Gemini 2.5
  16. The New Skill in AI is Not Prompting, It's Context Engineering
  17. Single vs Multi-Agent System?
  18. Zero to One: Learning Agentic Patterns
  19. Google Gemini LangChain Cheatsheet
  20. OpenAI Codex CLI, how does it work?
  21. Model Context Protocol (MCP) an overview
  22. ReAct agent from scratch with Gemini 2.5 and LangGraph
  23. Pass@k vs Pass^k: Understanding Agent Reliability
  24. Google Gemma 3 Function Calling Example
  25. Function Calling Guide: Google DeepMind Gemini 2.0 Flash
  26. From PDFs to Insights: Structured Outputs from PDFs with Gemini 2.0
  27. Mini-R1: Reproduce Deepseek R1 „aha moment“ a RL tutorial
  28. How to align open LLMs in 2025 with DPO and and synthetic data
  29. Bite: How Deepseek R1 was trained
  30. How to use Anthropic MCP Server with open LLMs, OpenAI or Google Gemini

2024 27

  1. Fine-tune classifier with ModernBERT in 2025
  2. How to fine-tune open LLMs in 2025 with Hugging Face
  3. Deploy QwQ-32B-Preview the best open Reasoning Model on AWS with Hugging Face
  4. Deploy Llama 3.2 Vision on Amazon SageMaker
  5. How to Fine-Tune Multimodal Models or VLMs with Hugging Face TRL
  6. Evaluate open LLMs with Vertex AI and Gemini
  7. Evaluate LLMs using Evaluation Harness and Hugging Face TGI/vLLM
  8. Deploy open LLMs with Terraform and Amazon SageMaker
  9. LLM Evaluation doesn't need to be complicated
  10. Evaluating Open LLMs with MixEval: The Closest Benchmark to LMSYS Chatbot Arena
  11. Train and Deploy open Embedding Models on Amazon SageMaker
  12. Deploy Mixtral 8x7B on AWS Inferentia2 with Hugging Face Optimum
  13. Fine-tune Llama 3 with PyTorch FSDP and Q-Lora on Amazon SageMaker
  14. Fine-tune Embedding models for Retrieval Augmented Generation (RAG)
  15. Understanding the Cost of Generative AI Models in Production
  16. Deploy Llama 3 70B on AWS Inferentia2 with Hugging Face Optimum
  17. Deploy open LLMs with vLLM on Hugging Face Inference Endpoints
  18. Efficiently fine-tune Llama 3 with PyTorch FSDP and Q-Lora
  19. Deploy Llama 3 on Amazon SageMaker
  20. Accelerate Mixtral 8x7B with Speculative Decoding and Quantization on Amazon SageMaker
  21. Deploy Llama 2 70B on AWS Inferentia2 with Hugging Face Optimum
  22. Fine-Tune and Evaluate LLMs in 2024 with Amazon SageMaker
  23. Evaluate LLMs with Hugging Face Lighteval on Amazon SageMaker
  24. How to fine-tune Google Gemma with ChatML and Hugging Face TRL
  25. RLHF in 2024 with DPO and Hugging Face
  26. How to Fine-Tune LLMs in 2024 with Hugging Face
  27. Scale LLM Inference on Amazon SageMaker with Multi-Replica Endpoints

2023 42

  1. Fine-tune Llama 7B on AWS Trainium
  2. Programmatically manage 🤗 Inference Endpoints
  3. Deploy Mixtral 8x7B on Amazon SageMaker
  4. Deploy Embedding Models on AWS inferentia2 with Amazon SageMaker
  5. Deploy Llama 2 7B on AWS inferentia2 with Amazon SageMaker
  6. Deploy Stable Diffusion XL on AWS inferentia2 with Amazon SageMaker
  7. Amazon Bedrock: How good (bad) is Titan Embeddings?
  8. Evaluate LLMs and RAG a practical example using Langchain and Hugging Face
  9. Deploy Idefics 9B and 80B on Amazon SageMaker
  10. Train and Deploy Mistral 7B with Hugging Face on Amazon SageMaker
  11. Llama 2 on Amazon SageMaker a Benchmark
  12. Fine-tune Falcon 180B with DeepSpeed ZeRO, LoRA and Flash Attention
  13. Fine-tune Falcon 180B with QLoRA and Flash Attention on Amazon SageMaker
  14. Deploy Falcon 180B on Amazon SageMaker
  15. Optimize open LLMs using GPTQ and Hugging Face Optimum
  16. LLMOps: Deploy Open LLMs using Infrastructure as Code with AWS CDK
  17. Deploy Llama 2 7B/13B/70B on Amazon SageMaker
  18. Introducing EasyLLM - streamline open LLMs
  19. Extended Guide: Instruction-tune Llama 2
  20. LLaMA 2 - Every Resource you need
  21. Fine-tune LLaMA 2 (7-70B) on Amazon SageMaker
  22. Train LLMs using QLoRA on Amazon SageMaker
  23. Deploy LLMs with Hugging Face Inference Endpoints
  24. Optimize and Deploy BERT on AWS inferentia2
  25. Securely deploy LLMs inside VPCs with Hugging Face and Amazon SageMaker
  26. Deploy Falcon 7B and 40B on Amazon SageMaker
  27. Fine-tune BERT for Text Classification on AWS Trainium
  28. Introducing the Hugging Face LLM Inference Container for Amazon SageMaker
  29. Generative AI for Document Understanding with Hugging Face and Amazon SageMaker
  30. How to scale LLM workloads to 20B+ with Amazon SageMaker using Hugging Face and PyTorch FSDP
  31. Setting up AWS Trainium for Hugging Face Transformers
  32. Train and Deploy BLOOM with Amazon SageMaker and PEFT
  33. Introducing IGEL an instruction-tuned German large Language Model
  34. Efficient Large Language Model training with LoRA and Hugging Face
  35. Deploy FLAN-UL2 20B on Amazon SageMaker
  36. Getting started with Pytorch 2.0 and Hugging Face Transformers
  37. Controlled text-to-image generation with ControlNet on Inference Endpoints
  38. Combine Amazon SageMaker and DeepSpeed to fine-tune FLAN-T5 XXL
  39. Fine-tune FLAN-T5 XL/XXL using DeepSpeed and Hugging Face Transformers
  40. Deploy FLAN-T5 XXL on Amazon SageMaker
  41. Hugging Face Transformers Examples
  42. Getting started with Transformers and TPU using PyTorch

2022 47

  1. Fine-tune FLAN-T5 for chat and dialogue summarization
  2. Managed Transcription with OpenAI Whisper and Hugging Face Inference Endpoints
  3. Stable Diffusion Inpainting example with Hugging Face inference Endpoints
  4. Stable Diffusion with Hugging Face Inference Endpoints
  5. Document AI: LiLT a better language agnostic LayoutLM model
  6. Multi-Model GPU Inference with Hugging Face Inference Endpoints
  7. Serverless Machine Learning Applications with Hugging Face Gradio and AWS Lambda
  8. Accelerate Stable Diffusion inference with DeepSpeed-Inference on GPUs
  9. Stable Diffusion on Amazon SageMaker
  10. Deploy T5 11B for inference for less than $500
  11. Outperform OpenAI GPT-3 with SetFit for text-classification
  12. Fine-tuning LayoutLM for document-understanding using Keras and Hugging Face Transformers
  13. Deploy LayoutLM with Hugging Face Inference Endpoints
  14. Document AI: Fine-tuning LayoutLM for document-understanding using Hugging Face Transformers
  15. Custom Inference with Hugging Face Inference Endpoints
  16. Accelerate GPT-J inference with DeepSpeed-Inference on GPUs
  17. Document AI: Fine-tuning Donut for document-parsing using Hugging Face Transformers
  18. Use Sentence Transformers with TensorFlow
  19. Pre-Training BERT with Hugging Face Transformers and Habana Gaudi
  20. Accelerate BERT inference with DeepSpeed-Inference on GPUs
  21. Accelerate Sentence Transformers with Hugging Face Optimum
  22. Deep Learning setup made easy with EC2 Remote Runner and Habana Gaudi
  23. Accelerate Vision Transformer (ViT) with Quantization using Optimum
  24. Optimizing Transformers for GPUs with Optimum
  25. Hugging Face Transformers and Habana Gaudi AWS DL1 Instances
  26. Optimizing Transformers with Hugging Face Optimum
  27. Convert Transformers to ONNX with Hugging Face Optimum
  28. Setup Deep Learning environment for Hugging Face Transformers with Habana Gaudi on AWS
  29. Static Quantization with Hugging Face `optimum` for ~3x latency improvements
  30. Advanced PII detection and anonymization with Hugging Face Transformers and Amazon SageMaker
  31. An Amazon SageMaker Inference comparison with Hugging Face Transformers
  32. Semantic Segmantion with Hugging Face's Transformers and Amazon SageMaker
  33. Automatic Speech Recogntion with Hugging Face's Transformers and Amazon SageMaker
  34. Serverless Inference with Hugging Face's Transformers, DistilBERT and Amazon SageMaker
  35. Accelerated document embeddings with Hugging Face Transformers and AWS Inferentia
  36. Save up to 90% training cost with AWS Spot Instances and Hugging Face Transformers
  37. Speed up BERT inference with Hugging Face Transformers and AWS Inferentia
  38. Creating document embeddings with Hugging Face's Transformers and Amazon SageMaker
  39. Autoscaling BERT with Hugging Face Transformers, Amazon SageMaker and Terraform module
  40. Multi-Container Endpoints with Hugging Face Transformers and Amazon SageMaker
  41. Asynchronous Inference with Hugging Face Transformers and Amazon SageMaker
  42. Deploy BERT with Hugging Face Transformers, Amazon SageMaker and Terraform module
  43. Task-specific knowledge distillation for BERT using Transformers and Amazon SageMaker
  44. Distributed training on multilingual BERT with Hugging Face Transformers and Amazon SageMaker
  45. Financial Text Summarization with Hugging Face Transformers, Keras and Amazon SageMaker
  46. Deploy GPT-J 6B for inference using Hugging Face Transformers and Amazon SageMaker
  47. Image Classification with Hugging Face Transformers and `Keras`

2021 12

  1. Workshop: Enterprise-Scale NLP with Hugging Face and Amazon SageMaker
  2. Hugging Face Transformers with Keras: Fine-tune a non-English BERT for Named Entity Recognition
  3. New Serverless Transformers using Amazon SageMaker Serverless Inference and Hugging Face
  4. Hugging Face Transformers BERT fine-tuning using Amazon SageMaker and Training Compiler
  5. MLOps: Using the Hugging Face Hub as model registry with Amazon SageMaker
  6. A remote guide to re:Invent 2021 machine learning sessions
  7. MLOps: End-to-End Hugging Face Transformers with the Hub and SageMaker Pipelines
  8. Going Production: Auto-scaling Hugging Face Transformers with Amazon SageMaker
  9. Deploy BigScience T0_3B to AWS and Amazon SageMaker
  10. Scalable, Secure Hugging Face Transformer Endpoints with Amazon SageMaker, AWS Lambda, and CDK
  11. Few-shot learning in practice with GPT-Neo
  12. Distributed Training: Train BART/T5 for Summarization using 🤗 Transformers and Amazon SageMaker

2020 19

  1. Multilingual Serverless XLM RoBERTa with HuggingFace, AWS Lambda
  2. Serverless BERT with HuggingFace, AWS Lambda, and Docker
  3. AWS Lambda with custom docker images as runtime
  4. New Serverless BERT with Huggingface, AWS Lambda, and AWS EFS
  5. efsync my first open-source MLOps toolkit
  6. My path to become a certified solution architect
  7. Create custom Github Action in 4 steps
  8. Fine-tune a non-English GPT-2 Model with Huggingface
  9. Mount your AWS EFS volume into AWS Lambda with the Serverless Framework
  10. Serverless BERT with HuggingFace and AWS Lambda
  11. How to use Google Tag Manager and Google Analytics without Cookies
  12. BERT Text Classification in a different language
  13. Scaling Machine Learning from ZERO to HERO
  14. Getting Started with AutoML and AWS AutoGluon
  15. K-Fold as Cross-Validation with a BERT Text-Classification Example
  16. How to Set Up a CI/CD Pipeline for AWS Lambda With GitHub Actions and Serverless
  17. Set up a CI/CD Pipeline for your Web app on AWS with Github Actions
  18. Getting started with CNNs by calculating LeNet-Layer manually
  19. Google Colab the free GPU/TPU Jupyter Notebook Service

Undated 2

  1. Philipp Schmid © 2026
  2. Imprint