Agentic AI
How to Evaluate Agents Locally with Arize Phoenix
A hands-on guide to running Phoenix locally: trace an agent over OpenTelemetry, then score it with LLM-as-judge and code evaluators.
7 min read
Folarin Akinloye is an AI Engineer based in London, UK. He builds production-ready agentic AI systems, multi-agent architectures, and sophisticated RAG implementations, and writes about the engineering decisions behind them.
Articles by Folarin
Agentic AI
What Is AgentOps? Operating AI Agents in Production
AgentOps is the practice of running agents in production: observability, evaluation, guardrails, cost governance, and human oversight.
6 min read
AI Infrastructure
What Is OpenTelemetry, and Why It Matters for Agent Observability
OpenTelemetry in plain terms: the data model, the GenAI semantic conventions, and why building agent tracing on OTel avoids lock-in.
6 min read
AI Infrastructure
Tracing in Agent Observability: What It Is and Why It Matters
What traces and spans actually are, how they map onto an agent run, and how to walk one to find the exact step that broke.
7 min read
LLM Integration
What the Research Actually Says About Prompting Reasoning
A tour of the reasoning research: how prompting elicits it, how researchers categorize the techniques, and why one prominent position paper says it is not reasoning at all.
4 min read
LLM Integration
A Field Guide to Model-Specific Prompting
Most prompting advice transfers across models. The parts that do not will quietly break your app when you switch providers. A working map of the differences.
5 min read
LLM Integration
A Real Prompt Engineering Case Study: 65.6 to 91.7 F1 on Job Classification
A production case study measured fourteen prompt modifications one by one. The results are humbling: naming the model helped, few-shot examples hurt, and templates cost accuracy.
4 min read
LLM Integration
Prompt Functions: Treating Prompts as Reusable Functions
Prompt functions wrap a prompt in a name, inputs, and a rule so you can call and compose it like code. The chat-window version is dated; the idea runs everything now.
6 min read
LLM Integration
Generating Code with Prompts: The Fundamentals Still Matter
Code generation is the most-used LLM capability and the most casually prompted. A tour of the core patterns, plus the checks that catch what the model quietly gets wrong.
6 min read
LLM Integration
Dataset Diversity: Fixing Repetitive Synthetic Generations
Cranking temperature does not make synthetic data diverse. Injecting randomness into the prompt does. Here is the technique, with code.
7 min read
Agentic AI
Building Agents That Actually Work
Most agents fail for boring, fixable reasons. Here are the practices that make an agent reliable: fewer LLM calls, tools that explain themselves, and a debugging ladder to climb when it misbehaves.
10 min read
RAG
Generating a Synthetic Dataset for RAG
When retrieval fails in your domain or language, you can prompt a large model to generate training data for a custom retriever, matching thousands of hand-labels for tens of dollars.
6 min read
LLM Integration
Generating Synthetic Data with Prompts
You can prompt a model to produce labeled examples on demand: test fixtures, eval sets, cold-start training data. Here is how to do it well and where the sharp edges are.
6 min read
LLM Integration
Bias in Prompting: How Your Prompt Design Skews the Model
Your few-shot examples carry hidden biases. The distribution and order of them can nudge the model toward a label before it reads the input. Here is how to spot it and fix it.
6 min read
LLM Integration
Factuality: Prompting to Reduce Hallucination
Models hallucinate because they are built to sound coherent, not to be right. Here are three prompt-level moves that reduce made-up answers, plus where prompting stops working.
6 min read
LLM Integration
Adversarial Prompting: Injection, Leaking, and Jailbreaking
Prompt injection, prompt leaking, and jailbreaking are three different attacks with three different fixes. Here is how each one works and what defence actually holds.
8 min read
LLM Integration
Graph Prompting, Explained
Graph prompting feeds a model structure, not just prose: entities as nodes, relationships as edges. Here is what the idea is, where it comes from, and when it pays off.
8 min read
Agentic AI
CodeAgent vs ToolCallingAgent: Two Ways to Let an Agent Act
smolagents ships two agent classes that act in completely different ways: one writes Python, one emits structured JSON. Here is how each works, what each is good and bad at, and when to reach for which.
9 min read
LLM IntegrationPrompt Engineering, Properly · Part 14
Prompting Reasoning Models Is Almost the Opposite of Prompting Chat Models
Reasoning models do the chain-of-thought for you. That flips a lot of prompting advice on its head, and getting it wrong wastes tokens and quality.
7 min read
LLM Integration
Multimodal Chain-of-Thought: Reason Over the Picture, Then Answer
Multimodal CoT extends step-by-step reasoning to text plus vision with a two-stage split: build a rationale from both modalities, then answer from the rationale.
5 min read
LLM Integration
PAL: Let the Model Reason in Words, but Let Python Do the Math
Chain-of-Thought lets a model reason and compute in the same breath, which is where it slips. PAL keeps the reasoning in language but hands the math to a Python interpreter.
6 min read
LLM Integration
Directional Stimulus Prompting: Train a Tiny Model to Whisper Hints to a Big One
Directional Stimulus Prompting trains a small model to produce hints that nudge a big frozen LLM toward what you want. It is prompt optimization when you cannot touch the weights.
5 min read
LLM Integration
Active-Prompt: Stop Guessing Which Few-Shot Examples to Annotate
Hand-picking chain-of-thought examples is a guess. Active-Prompt uses model disagreement to find the questions where a human annotation actually moves the needle.
6 min read
LLM Integration
ART: Let the Model Write Its Own Tool-Using Reasoning
ART made a frozen LLM generate its own multi-step reasoning, pause to call tools, and resume. It is 2023 research, but the pattern is everywhere in modern agents.
5 min read
LLM IntegrationPrompt Engineering, Properly · Part 7
Generated-Knowledge Prompting: Surface Facts Before You Answer
For questions that need world knowledge, having the model generate relevant facts first, then answer using them, beats answering cold. How it works, and when to use it over RAG.
6 min read
LLM IntegrationPrompt Engineering, Properly · Part 6
Self-Consistency: Sampling Your Way to Better Answers
One chain of thought can take a wrong turn. Self-consistency samples several reasoning paths for the same question and votes on the answer, beating a single greedy pass on reasoning tasks.
5 min read
LLM IntegrationPrompt Engineering, Properly · Part 5
Chain-of-Thought Prompting
Telling a model to reason before it answers genuinely improves accuracy on multi-step problems. How standard and zero-shot chain-of-thought work, when to use them, and the cost.
7 min read
LLM IntegrationPrompt Engineering, Properly · Part 4
Zero-Shot vs Few-Shot Prompting
Examples are the most effective context you can add to a prompt, until they aren't. When few-shot beats zero-shot, how many to use, how to pick them, and where examples won't help.
7 min read
LLM IntegrationPrompt Engineering, Properly · Part 3
The Anatomy of a Good Prompt
A good prompt has up to four parts: instruction, context, input data, and output indicator. Here is what each does and the design habits that make them land.
7 min read
LLM IntegrationPrompt Engineering, Properly · Part 2
The Settings That Change Your Output
Before you rewrite a prompt, check the dials around it. Temperature, top_p, max tokens, and the penalties each change your output in a specific way. Here is when to touch each.
6 min read
Agentic AI
The Agent Loop: Building ReAct From Scratch
An agent is a stateless model wrapped in a loop. Here is the ReAct pattern built from scratch: the message list as memory, the exact algorithm, and the rule that generate calls equal tool rounds plus one.
14 min read
AI Infrastructure
Git Worktrees Are Having a Moment Because of AI Agents
Git worktrees let you check out several branches into separate folders at once. They went from obscure to default because running coding agents in parallel needs isolated workspaces.
7 min read
Agentic AI
@Claude in Your Slack: What Claude Tag Actually Changes
Anthropic's Claude Tag drops a shared, persistent agent into Slack channels. The interesting part is not the model, it is the move from one-off chats to a teammate that remembers.
7 min read
AI Infrastructure
Monitoring Model Deprecation in Production
Why model deprecation quietly breaks production apps, and a concrete workflow: one models constant file, a scheduled GitHub Action checking deprecations.info plus a liveness probe, and an alert.
16 min read480 views
AI Infrastructure
Observability for LLM Apps: What to Log, What to Alert On
A practical guide to LLM observability: the traces and metrics to log, the handful of alerts worth waking up for, and why traditional monitoring misses what matters.
7 min read730 views
Agentic AI
Graph, State Machine, or Plain Loop: How to Structure an Agent
How to choose how to structure an agent: the plain while loop, the explicit state machine, and the full graph. What each gives you, what each costs, and when to move up.
8 min read540 views
AI Infrastructure
Caching Agent Tool Calls (Not Just Prompts)
Prompt caching is solved. The bigger win for agents is caching tool-call results, and that is harder because tools have side effects. Here is how to do it safely.
9 min read290 views
RAG
Building a Multi-Tenant RAG System: Isolation, Per-Tenant Indexes, and the Leaks Nobody Plans For
How to build a RAG system that serves many customers without leaking data between them: silo vs pool vs bridge, namespaces vs metadata filters, and where isolation must hold.
9 min read620 views
Agentic AI
Streaming Tool Calls and Partial Outputs in LangGraph
How to stream LLM tokens, tool progress, and state updates from a LangGraph agent using the v2 stream API, with the custom writer pattern most people miss.
9 min read410 views
AI Infrastructure
Five Projects to Actually Master AI Engineering (for Experienced Engineers)
If you can already ship software, the way into AI engineering is not another tutorial. It is building five systems that force you to handle failure, measure quality, and run things in production.
7 min read13.8K views
Agentic AIBuilding with DeepAgents · Part 5
Human-in-the-Loop in DeepAgents: Approve Before It Acts
Autonomy is great until the agent deletes the wrong file or emails the wrong list. Here is how DeepAgents pauses on sensitive tool calls and waits for a human, in TypeScript.
11 min read9.5K views
Agentic AIBuilding with DeepAgents · Part 4
Why Skills Matter When You Build an Agent
Skills give an agent deep, reusable capability without paying for it on every turn. Here is how DeepAgents loads them in layers, and how to write ones that get picked.
10 min read15.7K views
Agentic AIBuilding with DeepAgents · Part 3
Memory in DeepAgents: How Agents Learn Across Conversations
DeepAgents treats long-term memory as files the agent reads and writes, with backends deciding where they live. Here is how scoping, episodic memory, and consolidation actually work.
9 min read8.1K views
Agentic AIBuilding with DeepAgents · Part 2
Subagents in DeepAgents: Delegation That Keeps Context Clean
Subagents are the main lever for stopping a long task from drowning in its own context. Here is how DeepAgents implements them in TypeScript, and the mistakes that make them useless.
10 min read10K views
Agentic AIBuilding with DeepAgents · Part 1
Context Engineering in DeepAgents, From the Inside
A tour of how DeepAgents assembles, compresses, and isolates context, with the TypeScript config for each layer. The framework does a lot automatically; this is what is happening and where you steer it.
12 min read8.9K views
LLM Integration
Prompt Caching for LLM Apps: What It Is and When It Pays Off
Prompt caching can cut input cost by up to 90%, but only if you structure prompts so the cache actually hits. Here is how it works and when it pays off.
6 min read7.3K views
Agentic AI
LangGraph State, Checkpointing, and Persistence Explained
What state, checkpoints, and threads actually are in LangGraph, how to pick a checkpointer, and how persistence powers memory, resume, and human-in-the-loop.
7 min read6.3K views
LLM Integration
Cutting LLM Cost and Latency Without Wrecking Quality
A practical playbook for making LLM apps cheaper and faster without hurting answers: where the cost and latency actually go, and the levers that work.
7 min read15.4K views
Agentic AI
Guardrails and Safety for Agents in Production
A practical guide to keeping production agents safe: where to put input and output rails, how to contain tools, and why prompt injection needs layers, not a fix.
7 min read21.4K views
LLM Integration
Streaming LLM Responses End to End: Backend to UI
A full walkthrough of streaming LLM output from the provider, through your backend, into a React UI: SSE, backpressure, cancellation, and the gotchas.
9 min read6.2K views
AI Infrastructure
How I Learn to Build Production AI Systems by Dissecting Open Source
The field moves faster than any course can. So I learn production AI by gutting and rebuilding the open-source projects that production teams actually run. Here is the path.
7 min read11.6K views
Agentic AI
Inside Vercel's Eve: A Deep Technical Walkthrough, with a Research Agent
A deep, hands-on tour of Vercel's Eve framework for durable agents: the filesystem-first model, every core capability, a complete TypeScript research agent, and an honest comparison with LangChain's Deep Agents.
27 min read11.9K views
Agentic AIAgentic AI in Practice · Part 7
Agent Memory: Short-Term vs Long-Term, and How to Wire It Up
A clear split between an agent's working memory and its cross-session memory, with LangGraph code for checkpointers, stores, and the semantic/episodic/procedural breakdown.
7 min read711 views
LLM Integration
Structured Outputs and Function Calling, In Depth
Structured outputs guarantee the model's response matches your JSON schema. Here is how it works under the hood, how it relates to function calling, and the patterns that hold up in production.
7 min read14.1K views
AI InfrastructureProduction RAG · Part 4
Choosing a Vector Database in 2026: pgvector vs Pinecone vs Qdrant vs Weaviate
A practical, opinionated comparison of the four vector databases people actually argue about, with a clear default and the signals that mean you have outgrown it.
6 min read8.2K views
RAGProduction RAG · Part 3
Embeddings Explained for Engineers
Embeddings turn text into vectors so meaning becomes geometry. Here is the mental model, the gotchas, and how to choose a model in 2026.
5 min read10.1K views
RAGProduction RAG · Part 2
Reranking in RAG: Cross-Encoders and When They Are Worth the Latency
Rerankers re-score your top candidates with a model that actually reads the query and the document together. Here is how they work and when they pay off.
5 min read1.8K views
RAGProduction RAG · Part 1
Chunking Strategies for RAG: Fixed, Recursive, Semantic, and How to Choose
Chunking quietly decides how good your RAG system can ever be. Here is how the main strategies compare and a simple way to pick one.
6 min read6.2K views
Agentic AIAgentic AI in Practice · Part 6
Deep Agents by LangChain: A Deep Dive, with a Customer Support Agent
Deep Agents bakes planning, context management, subagents, and human approval into the agent loop. Here is why it exists, every main feature, and a production-grade customer support agent built with it.
16 min read15K views
Agentic AIAgentic AI in Practice · Part 5
Evaluating Agents with LangSmith: A Complete Guide
You cannot improve an agent you cannot measure. A deep, practical guide to evaluating agents with LangSmith: tracing, datasets, evaluators, experiments, and what makes agent evaluation different.
11 min read12.2K views
Agentic AIAgentic AI in Practice · Part 4
Building MCP Servers with FastMCP: A Practical Guide
A hands-on guide to building MCP servers with FastMCP, drawn from two I actually shipped: the anatomy, the rules, stdio vs remote HTTP, the pitfalls, and how to test and deploy.
9 min read4.9K views
Agentic AIAgentic AI in Practice · Part 3
Giving Agents Tools: Function Calling and MCP, Explained
An agent with no tools is just a chatbot. Here is how tool calling really works, what makes a tool the model can use reliably, and how MCP lets you plug in tools you did not write.
8 min read16.1K views
RAGAgentic AI in Practice · Part 2
Agentic RAG, and How It Differs from Traditional RAG
Traditional RAG retrieves once and hopes for the best. Agentic RAG lets the model decide whether, what, and how to retrieve, then check its own work. Here is the difference, with code.
8 min read19.1K views
Agentic AIAgentic AI in Practice · Part 1
What Are AI Agents, and What Is a Multi-Agent System?
Agents, explained without the hype: what they are, when one is enough, when you need several, and how to build a working multi-agent system with LangChain.
10 min read2.9K views