-
Code mode for mcp servers and llms
Code mode for mcp servers is about LLM writing and calling the code to use a proper MCP instead of calling it directly with the whole context. It makes the call a lot smaller, no overhead is passed, just the basics that are required to call the proper MCP method. Just as You do in code, method or a function, proper parameters, everything validated and… BAM ! We are returning a context that LLM uses in further stages. Anthropics wrote… This is really nice looking but only for a huge models with a 1kk tokens of a context. We need to remember that this is not possible on any kind…
-
Data contracts. Building universal data access proxy api
Data contracts usually requires us to build around them a universal data access proxy API for users to consume. API utilizing proper data contracts, negotiated with different teams, acts as a unified gateway providing the necesities. Allows access to data sources like databases, REST APIs, GraphQL endpoints or other file systems. One api to rule them all Steps to build data contracts for proxy api You could try and adopt a similar flow for creating such access points, even make a template in JIRa so You will know where to get proper data and how to acquire it… or maybe expose the library and just aprove properly looking merge requests……
-
97 Things Every Data Engineer Should Know – book review
97 Things Every Data Engineer Should Know review will be a positive one. This style of books is currently my favorite. Might get another one from the series 🙂 It is bits and pieces of knowledge You can digest easily. Scattered across multiple disciplines teachings are of a principle design. Book is technologically agnostic, meaning rules, law, principles and methodologies presented You can use with any framework or system. It is like the design principles. Great read, would recommend. 97 lessons to pick from The book is all about best practices, system design, queues, asynchronouse communication and many more. You can easily read it day by day when You „meditate”…
-
Do not be a hero – 3 pros and some tradeoffs
Do not be a hero. Why ? . Ever seen that one guy doing midnight fixes, hot releases after hours and then gets praised in front of everyone making You look bad ? When features are delivered on time and on target the team is considered heroes. We all love rewards and recognition but heroism might become … Heroes give up work / life balance. As the ancient wisdom teaches us „no good deed goes unpunished”. Whenever there is a new crisis or a target to achieve is set, yesterday’s heroes are quickly forgotten. Long hours eventually might lead to burnout, anxiety and/or even depression. What “Don’t Be a Hero”…
-
5 why methodology
5 why methodology is one of my favorites. Simple, easy to remember and always work. If You are not at Your root cause just ask… yet again… Why ! Remember when you burned that pork chop in the kitchen last time and instead of blaming the recipe, you dug deeper? „Why did it burn? Fire too high. Why? Didn’t check the stove. Why? Rushed after back to the pc…” Boom! You found the reason. Do not ever leav e a frying pan unattended or start using a timer. That’s 5 whys methodology in action, a powerhouse technique from Toyota smashing problems and stripping the mystery Basics to remember 5 Why…
-
When to use Value Objects in database
Question when to use value objects in database is usually discussed during storing a new set of data. Usage of value objects should be considered while modeling abstractions that are constant / immutable and do not require unique identity. Examples can be: geospatial data, estate addresses or any categories that belong to an entity. VO also encapsulates validation logic and business rules. Helps to avoid raw strings or magic numbers. When to use it and its benefits How to map a value object in db Scenario Mapping Benefit Embed Columns in parent table (Street, City in Address) Fast reads, no joins Complex nesting JSONB/PostgreSQL or owned entities Flexible schema, validation…
-
Law of Demeter in programming
Law of Demeter in programming You could in short paraphrase as „Don’t talk to strangers!” or „Keep Your hands in your pockets”. A class needing data from a deep chain of objects like – user.getAddress().getStreet().getNumber().It is visible very often in OOP, less in functional approachess ! The Law of Demeter states that method only knows its closest dependencies: itself, passed parameters, created objects, and own fields. Should not be used with dot chains ( they were so fancy when we started Java 8 back in the day). What to Remember? Where Law of Demeter comes from ? Why Demeter ? The Law of Demeter in progamming takes its name from…
-
How to use Context7 mcp server
What is Context7? Context7 is an open-source MCP server that provides real-time, version-specific documentation and code examples for over 50,000 libraries. It integrates seamlessly with AI tools like GitHub Copilot, Claude, or custom LLM agents. Why use it? How to use Context7 mcp server ? You can play around on the main page of the project or simply use a curl to fetch the data You wish so look up : context7.com/api/v2/docs/code/vuejs/pinia?topic=log&tokens=666 How to Configure Context7 How to use Context7 is really simple, just add a proper config entry and the plugin You use should pick it up instantly. Maybe reload the app if in need. 1. Install the Context7…
-
Tokenization and embedding of song lyrics „We will, we will…”
Tokenization and embedding of song lyrics „We will, we will…” i know you know how it ends. but have You wondered what would an LLM say ? Let us find out. I want to ask the Clause Sonnet 4.5 about the embeddings, tokenizations and probability of figuring out the lyrics for „We will, we will…” prompt 🙂 Can you show me the tokenization, embeddings, emtadata and probabilities for „We will, we will …” Tokenization Any kind of machine learning uses numbers in the back stage. The text is split into tokens, each mapped to an ID ( example values ) : Token ID Token Text Type Position Length 1234 We…
-
What is a token in AI ? Prompt examples
What is a token in AI ? – A piece of text, usually a word that we send to the LLM. The same goes for when we get a response ( usually number of words = number of tokens, roughly). How to use it ? Since we pay for it , rather cautiously. We want to send the minimum and get the maximum out of every request. Pretty much the basics of economics. Below are couple of prompts i used to analyze my tokens usage. Something are obsiously hallucinations but on the other hand we get a prety decent breakdown of all the data i did send for that coding…
