1290 posts · Page 79 of 108. ← Blog
Four protocols, four jobs: how tools, agent-to-agent messaging, and payment rails compose into one buying flow.
How the layers compose: MCP gives an agent tools and context; A2A lets agents talk to each other; AP2/ACP/x402 are the payment rails those agents call when money must move.
How to build PSD2-grade open-banking APIs — strong customer authentication and its exemptions, the consent grant and its revocation, and the hard boundary between account-information and payment-initiation scopes.
Teaches how to build open-banking APIs: strong customer authentication and exemptions, consent grant/lifecycle, account-information vs payment-initiation scopes, and TPP authorization with token/consent revocation.
How to turn a book of exposures into risk-weighted assets and a capital ratio with a deterministic, auditable pipeline
Teaches the engineering of a risk-weighted-asset calculator: exposure classification, risk-weight lookup (standardized approach), credit-conversion factors for off-balance items, and capital-ratio aggregation for regulatory reporting.
Building low-latency sliding-window counters, entity-keyed aggregates, and an online feature store that stays consistent with its batch-computed twin.
Teaches how to build low-latency velocity checks: sliding-window counters and aggregates in an online feature store, entity keys (card/device/IP), and consistency between real-time and batch-computed features.
How a plain function becomes a callable tool, how ToolContext reaches session state, and how long-running tools pause a run for a human.
Tools give agents capabilities: a plain function becomes a tool with its signature as the schema, plus ToolContext, built-in tools, and long-running/human-in-the-loop tools across Python and Go.
How to pause a workflow to ask a human, checkpoint every super-step, then rewind the whole graph to a saved checkpoint and replay.
A RequestPort pauses to ask a human for a guess while every super-step is checkpointed, then RestoreCheckpoint rewinds the whole graph including tries state.
The graph model underneath every multi-agent app: executors as nodes, edges as data flow, and typed events streaming out as it runs.
The Microsoft Agent Framework workflow model in Python: executors as nodes, edges as data flow, switch-case routing, and typed streaming events - learned model-free.
Almost nobody trains a computer-vision model from scratch — and if you're trying to, you're probably doing it wrong. The real practice of vision is transfer learning: take a model pretrained on millions of images, adapt it to your task with a fraction of the data. This, plus smart data strategy, is how vision gets built in the real world, and it's the single most practical thing to understand about applying computer vision. This post is about how it's actually done.
Almost nobody trains vision models from scratch — the real practice is transfer learning: adapt a pretrained model with a fraction of the data. Feature extraction vs fine-tuning (chosen by data size), and the data realities (labeling cost, representativeness, augmentation) that actually decide a vision project.
The demo is where technical sales succeeds or fails — and where the most common, most fixable mistake happens. Given a chance to show the product, the enthusiastic builder gives the full tour: every feature, every capability, in order. The customer, who came with one specific problem, sees a wall of things that don't obviously address it, and disengages. A great demo does the opposite: it shows how the product solves this customer's problem, and almost nothing else. Getting demos and proofs of concept right is one of the highest-leverage skills in technical selling.
The demo is where technical sales succeeds or fails — and where the most common, fixable mistake happens: the enthusiastic builder gives the full feature tour, and the customer, who came with one problem, disengages. A great demo shows how the product solves this customer's problem, and almost nothing else. Demos and POCs are among the highest-leverage skills in technical selling.
Why Python draws a hard line between text and raw bytes, how the encode/decode boundary works, and the string-handling habits that keep that line from cutting you.
Why Python draws a hard line between text and raw bytes, how the encode/decode boundary works, and the string-handling habits that keep that line from cutting you.
Go's control flow is deliberately small — one loop keyword, a switch that doesn't fall through, an `if` that can scope its own variable — and then there's `defer`, the one construct that repays close reading. A tour of the whole surface, with the sharp edges labelled.
Go's control flow is deliberately small — one loop keyword, a switch that doesn't fall through, an `if` that can scope its own variable — and then there's `defer`, the one construct that repays close...
Why "authorize then capture" is a lie, and how to model a hold that grows, shrinks, expires, and reconciles against a moving available balance.
Pre-auth vs estimated vs incremental authorization (hotels, fuel, delivery), partial approvals, and reconciling holds against final capture.