Memory-Aware Agent Harnesses for Anti-Money Laundering — Nacho Martinez, Oracle
This workshop explains how an agent harness supplies the memory, perception, tools, context engineering, and control loop that a language model lacks on its own. An anti-money-laundering scenario grounds design choices such as files versus databases, short- versus long-term memory, context compaction, and selective retrieval of skills and tools.
AI EngineerWatchTranscript found
Quick learning frame
Read this before watching.
A context/search lesson is about getting the right evidence into the agent at the right time through indexes, search, memory, or knowledge graphs.
New playlist item from AI Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to design a memory-aware agent harness that assembles only the relevant evidence, tools, skills, and prior experience needed for a long-running decision workflow.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01Work question
02Source inventory
03Index/search layer
04Retrieval rule
05Agent context
06Answer/proof
07Maintenance
Deep lesson
Turn this video into working knowledge.
8,605 cleaned transcript words reviewed across 2,440 timed caption segments.
Thesis
Memory-Aware Agent Harnesses for Anti-Money Laundering — Nacho Martinez, Oracle teaches a practical context/search move: This workshop explains how an agent harness supplies the memory, perception, tools, context engineering, and control loop that a language model lacks on its own. An anti-money-laundering scenario grounds design choices such as files versus databases, short- versus long-term memory, context compaction, and selective retrieval of skills and tools.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:12
Design the Harness
“Hi. Hi everyone. Uh thank you for coming to this session. Uh we're it's going to be a session about agent harnesses. We don't have a lot of time. It's only one hour session. So um hopefully we...”
A frontier model supplies reasoning, but the harness controls the agent's memory, tool use, perception, context engineering, and repeated action loop. For anti-money-laundering work, those components let a bank flag suspicious transaction patterns for human review without expecting the model alone to sustain hours of autonomous work. Sketch an AML agent stack and label which component supplies reasoning, memory, transaction inputs, tools, and the loop that decides the next action.
16:32
Engineer Durable Memory
“short-term versus long-term memory. And in this case, this is part two. We have some short-term memory components. For example, if we are talking with clot or with codeex, we have short-term memory in the sense that the...”
The context window is short-term memory, while long-term memory can be separated into episodic records of events, procedural skills and workflows, and semantic facts or preferences. Because larger contexts degrade performance, a harness should summarize conversations or compact chunks into external storage that can be recalled by ID when needed. Classify five pieces of agent information as episodic, procedural, or semantic, then decide which should remain in the active context and which should be compacted.
36:39
Retrieve Context Selectively
“you're interact with interacting with a frontier AI model, you're essentially making the whole thing through an HTTP request and everything like this would be for example 10K or 5K as a system prompt. you're an AI assistant...”
The agent loop dynamically assembles context from memory, files, databases, tools, and skills before each reasoning call. In the workshop design, separate vector stores retrieve up to three relevant tools and three relevant skills, with an optional reranker narrowing them further so the prompt carries only what the task needs. Define a retrieval policy for one AML investigation that names the tool and skill stores, the top-k limit, and the condition for reranking to a single item.
01
Work question
Start with this video's job: This workshop explains how an agent harness supplies the memory, perception, tools, context engineering, and control loop that a language model lacks on its own. An anti-money-laundering scenario grounds design choices such as files versus databases, short- versus long-term memory, context compaction, and selective retrieval of skills and tools. Treat "Work question" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:12, where the video says: “Hi. Hi everyone. Uh thank you for coming to this session. Uh we're it's going to be a session about agent harnesses. We don't have a lot of time. It's only one hour session. So um hopefully we...”
02
Source inventory
Use "Source inventory" to locate the part of the context/search mechanism the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 16:32, where the video says: “short-term versus long-term memory. And in this case, this is part two. We have some short-term memory components. For example, if we are talking with clot or with codeex, we have short-term memory in the sense that the...”
03
Index/search layer
Turn "Index/search layer" into the reusable artifact for this lesson: A context retrieval map with source inventory, indexing/search path, query rules, freshness checks, and agent handoff. This is where watching becomes something you can inspect and reuse.
04
Retrieval rule
Use "Retrieval rule" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Agent context
Use "Agent context" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Answer/proof
Use "Answer/proof" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
07
Maintenance
Connect "Maintenance" to Memory-Aware Agent Harnesses for Anti-Money Laundering — Nacho Martinez, Oracle by naming the claim, the evidence, and the artifact it should produce.
Example
Source-backed artifact packet
Convert the video into a scoped artifact request that includes the transcript claim, mechanism, acceptance criteria, and proof. The output should be a context retrieval map with source inventory, indexing/search path, query rules, freshness checks, and agent handoff..
Example
Context/search proof brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the context/search pattern.
Example
Teach-back module
Transform the lesson into a definition, a Work question -> Source inventory -> Index/search layer -> Retrieval rule -> Agent context -> Answer/proof -> Maintenance diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
dumping all context
stale memory
retrieval with no proof trail
Letting the lesson drift into generic context-window advice.
Letting the lesson drift into memory hype without retrieval rules.
Letting the lesson drift into source claims without freshness checks.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: This workshop explains how an agent harness supplies the memory, perception, tools, context engineering, and control loop that a language model lacks on its own. An anti-money-laundering scenario grounds design choices such as files versus databases, short- versus long-term memory, context compaction, and selective retrieval of skills and tools.
02
Explain the practical stakes without hype: New playlist item from AI Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Work question -> Source inventory -> Index/search layer -> Retrieval rule -> Agent context -> Answer/proof -> Maintenance sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A context retrieval map with source inventory, indexing/search path, query rules, freshness checks, and agent handoff.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: Memory-Aware Agent Harnesses for Anti-Money Laundering — Nacho Martinez, Oracle
- URL: https://www.youtube.com/watch?v=KtBivPzmOM8
- Topic: Agentic Engineering
- My current learning frame: Design a small AML harness that flags one suspicious transaction pattern, separates its three long-term memory types, and assembles a compact prompt from the most relevant stored evidence, tool, and skill.
- Why this matters: New playlist item from AI Engineer; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:12 / Evidence 1: "Hi. Hi everyone. Uh thank you for coming to this session. Uh we're it's going to be a session about agent harnesses. We don't have a lot of time. It's only one hour session. So um hopefully we..."
- 4:12 / Evidence 2: "the loop until they arrive to uh to the conclusion or the answers, right? So the agents themselves don't have the capacity to do this. Why? Because I mean the context window is limited. The memory components are..."
- 8:18 / Evidence 3: "these systems are a combination of three and four which is um you know a combination of LLM driven workflows that provide automation and AI agents that provide autonomy. So it's a combination of automation and autonomy and..."
- 10:48 / Evidence 4: "harness is essentially all the scaffolding that you need to do to build uh these three components which is the the tool use the memory and the perception and we're not going to focus too much on the..."
- 16:32 / Evidence 5: "short-term versus long-term memory. And in this case, this is part two. We have some short-term memory components. For example, if we are talking with clot or with codeex, we have short-term memory in the sense that the..."
- 26:26 / Evidence 6: "we can pass to a model. The most important one is the system prompt, right? Everyone knows about this the system prompt concept. Um, and the system prompt is fine, but it cannot hold everything because otherwise things..."
- 36:39 / Evidence 7: "you're interact with interacting with a frontier AI model, you're essentially making the whole thing through an HTTP request and everything like this would be for example 10K or 5K as a system prompt. you're an AI assistant..."
Video-aware target:
- Prompt lane: Context/search
- Mechanism to extract: Extract how context is found, filtered, refreshed, and handed to the agent before it acts.
- Artifact to produce: A context retrieval map with source inventory, indexing/search path, query rules, freshness checks, and agent handoff.
- Artifact must include: source inventory; index/search layer; query rule; freshness check; agent handoff; proof behavior
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Extract how context is found, filtered, refreshed, and handed to the agent before it acts. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A context retrieval map with source inventory, indexing/search path, query rules, freshness checks, and agent handoff.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Work question -> Source inventory -> Index/search layer -> Retrieval rule -> Agent context -> Answer/proof -> Maintenance
- answers to these source questions: What source is searched or indexed? | What query/retrieval rule is demonstrated? | How does the agent use the retrieved context?
- 3 concrete examples that apply the video idea to real agentic work, such as codebase memory; personal wiki retrieval; Elastic search context engineering
- 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: dumping all context; stale memory; retrieval with no proof trail
- a checklist for the next real workflow, focused on: sources, query, freshness, handoff, citation/proof
- one practical exercise with a clear done signal: Write three retrieval queries for one real project and define what each must return.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "Memory-Aware Agent Harnesses for Anti-Money Laundering — Nacho Martinez, Oracle", not a generic Agentic Engineering essay.
- Cite the transcript wherever the prompt names a task boundary, review habit, context move, or verification standard.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: generic context-window advice; memory hype without retrieval rules; source claims without freshness checks.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
Agentic engineering means letting agents do everything.
It means designing work so agents can do bounded pieces well.
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a context retrieval map with source inventory, indexing/search path, query rules, freshness checks, and agent handoff..
A reusable artifact with a done signal and one verification step.03
Context/search teach-back card
Explain the context/search mechanism to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
Which agent capabilities belong to the harness rather than the frontier model's reasoning core?
How do episodic, procedural, and semantic long-term memory differ?
How does the workshop keep tools and skills from overwhelming the active context?
Source shelf
Use the video as a doorway, then verify with primary sources.
