InTowards AIbyNicholaus Lawson·May 23I Added 50 Tools to My AI Agent, and It Got Dumber.I built ToolGate, an MCP proxy that uses semantic search to filter AI agent tools. It reduces context bloat by 60%+ while maintaining 85%+…A response icon1A response icon1
InTowards AWSbyNicholaus Lawson·Feb 5An AWS-Native Pattern for Perplexica: SearXNG, LiteLLM, and BedrockUse ECS Fargate, self-hosted SearXNG, and an OpenAI-compatible LiteLLM proxy to run Perplexica with Bedrock models like Claude and Nova —…
InTowards AWSbyNicholaus Lawson·Sep 18, 2025Deploy N8N on AWS ECS Fargate Spot Instances for Workflow AutomationStep-by-step guide to Deploying N8N on AWS ECS Fargate with Spot Instances for Scalable, Cost-Saving Workflow Automation. Complete with…
InTowards AWSbyNicholaus Lawson·Aug 13, 2025From CLI to Conversational GenAIHow to Integrate AWS Q Dev CLI with Amazon Bedrock Agents and Knowledge Bases
InTowards AWSbyNicholaus Lawson·Mar 27, 2025Inside the AWS LLM League: Lessons from Fine-Tuning LLMs for a City’s 311 ServiceLeveraging Amazon SageMaker Jumpstart to fine-tune Llama 3.2 8B Instruct model to power a chatbotA response icon1A response icon1
InLevel Up CodingbyNicholaus Lawson·Mar 19, 2025Advanced Error Budgets in SREA Tiered Approach to Reliability
InLevel Up CodingbyNicholaus Lawson·Mar 3, 2025Additive-Only Programming Paradigm Part 1Introducing A Sustainable Approach to Software Evolution
InData Science CollectivebyNicholaus Lawson·Feb 27, 2025Fine-Tuning vs. Distillation. When to use eachUnderstanding why you need them and when to choose one versus using a base modelA response icon1A response icon1
InTowards AWSbyNicholaus Lawson·Nov 9, 2024Building Model-Agnostic GenAI ApplicationsUsing LiteLLM as a Proxy for Amazon BedrockA response icon1A response icon1
InAI AdvancesbyNicholaus Lawson·Jan 17, 2024LLMs in Financial Services — a money manager in your pocketIncluding how, why, and what pitfalls to be aware of when asking LLMs to answer questions about your money