Harness engineering: how we build with agents
Inspect a tenant-scoped export: its contract, observed output and verification record.
Read the guideInsights & Case Studies
Engineering perspectiveson building with agents and operating agent systems.
Methods, inspection guides and selected work. How software gets built with agents, and what changes when the product itself acts.
2 Case Studies
5 Articles
27 Guides
6 Tools & examples
Start here
The method and the questions to test it.
What to inspect in an agent-built codebase
A practical review of behavior, permissions, maintainability and handoff. Free to read.
Read the guideFor engineering leaders
Make the next decision with evidence.
Practical guides to choosing, funding and operating an AI system. Open frameworks, worked examples and no signup gate.
Vendor selection
How to choose an AI agent development company
An evidence-based vendor scorecard for CTOs and founders: compare architecture, evaluation, security, delivery and the system you will own.
Budget and economics
AI agent development cost: a practical planning model
Model the scope, operating costs and cost per successful task of an AI agent. Includes a transparent calculator and a worksheet for comparing proposals.
Architecture decisions
Build vs. buy AI agents: decide what you need to own
A layer-by-layer framework for enterprise AI agent decisions, with a comparison matrix, pilot acceptance criteria and an exit test.
40 items
Guide
10 min read
How to choose an AI agent development company
An evidence-based vendor scorecard for CTOs and founders: compare architecture, evaluation, security, delivery and the system you will own.
Guide
9 min read
AI agent development cost: a practical planning model
Model the scope, operating costs and cost per successful task of an AI agent. Includes a transparent calculator and a worksheet for comparing proposals.
Guide
10 min read
Build vs. buy AI agents: decide what you need to own
A layer-by-layer framework for enterprise AI agent decisions, with a comparison matrix, pilot acceptance criteria and an exit test.
Guide
9 min read
AI agent evaluation: a release framework
Define task success, test permissions and side effects, evaluate failure slices, and turn agent evaluations into an inspectable release decision.
Guide
8 min read
MCP server security: a review checklist
A practical MCP security review covering identity, authorization, tool execution, prompt injection, state handles and evidence for a release.
Guide
9 min read
AI technical due diligence: what the evidence needs to show
A technical diligence framework for AI software investments and vendor decisions: request the right evidence, test material claims and prioritize findings.
Guide
6 min read
AI agents vs. workflows: choosing the right control model
Use a workflow when the path is known. Introduce an agent where runtime judgment earns its complexity—and keep consequential actions under explicit control.
Guide
6 min read
AI agent architecture: a reference design
A reference architecture for identity, state, tool execution, permissions, evaluation and recovery—with a concrete workflow a CTO can inspect.
Guide
5 min read
Integrating AI into an existing enterprise platform
Add AI capabilities through existing identity, domain APIs and data contracts. Plan the migration, rollout and recovery before introducing autonomous writes.
Guide
6 min read
MCP vs. APIs: where each belongs
Compare MCP and direct APIs for discovery, domain behavior, permissions and execution. Decide when a shared protocol interface is useful.
Guide
5 min read
Single agent vs. multi-agent systems
Compare single-agent and multi-agent designs through task evidence, coordination cost, context handoff and responsibility for the final action.
Guide
5 min read
RAG vs. fine-tuning: diagnose the missing capability
Choose retrieval or fine-tuning by diagnosing missing knowledge and behavior. Compare freshness, access, evaluation evidence and operating requirements.
Guide
5 min read
Taking an AI prototype to launch
Turn a useful demo into an operable product with explicit task outcomes, evaluation, rollout gates, recovery and ownership.
Guide
5 min read
AI agent observability: from traces to decisions
Trace task outcomes, tool actions and waiting states to explain failures, control cost and investigate incidents while limiting sensitive data collection.
Guide
6 min read
RAG evaluation: test retrieval, answers and access
Separate retrieval quality, answer support and permission failures. Use a small worked corpus to understand what each metric establishes—and what it misses.
Guide
5 min read
LLM as a judge: calibration before automation
Use model graders for tasks they can judge reliably. Calibrate against reviewed examples, inspect missed failures and keep critical release rules explicit.
Guide
5 min read
LLM model routing and fallback
Choose model routes by task evidence, data policy and operating limits. Treat fallback as a tested execution path—not permission to send any workload anywhere.
Guide
5 min read
LLM latency optimization for useful completion
Locate delays across queues, models, tools and approval. Compare latency changes against task quality, cost and reliable completion.
Guide
6 min read
AI agent authorization at the action boundary
Define who can do what to which resource, derive identity from trusted context and recheck authority when an agent action executes.
Guide
5 min read
Prompt injection defenses for tool-using agents
Review prompt injection defenses at the data and tool boundaries, including scoped access, output validation, approval and adversarial tests.
Guide
5 min read
Human approval workflows for AI agents
Bind human approval to a specific AI agent action. Review scope, expiry, changed state and the authorization needed before execution.
Guide
5 min read
Permission-aware RAG: access before retrieval
Preserve tenant and source permissions through retrieval, context assembly, caching and citations. Treat revocation and index freshness as part of the design.
Guide
5 min read
Idempotent AI tool actions and safe retries
Design retries for AI tool actions with operation identity, transactional boundaries and reconciliation. Inspect the actual business effect.
Guide
5 min read
AI agent ROI: build a defensible business case
Assess AI agent ROI through adoption, remaining effort, useful capacity and operating costs. Design a pilot around the assumptions that matter.
Practical tool
5 min read
AI ROI calculator: capacity, cost and payback
Estimate workflow value, adoption, remaining effort and payback using your own inputs. A private browser worksheet with formulas and explicit limitations.
Practical tool
4 min read
AI evaluation sample planner
Calculate a one-sided zero-failure binomial bound and the trials needed for a target. Check the statistical assumptions before using the result.
Practical tool
5 min read
AI vendor evaluation scorecard
Compare two engineering proposals with fixed criteria, evidence confidence and separate critical requirements. Keep the worksheet private and export it locally.
Engineering example
5 min read
Engineering example: a retry after a lost response
Run a transaction example that commits once, loses the response and recovers the result. Inspect duplicate, conflict and tenant-scope behavior.
Engineering example
5 min read
Engineering example: permissioned retrieval
Run a synthetic retrieval example with tenant and document access checks. Inspect permitted context, denied requests and the boundaries still untested.
Engineering example
4 min read
Engineering example: an evaluation release gate
Inspect a runnable evaluation release gate that separates average task quality from critical failures, with fixtures you can run yourself.
Case study
2 min read
Agent Platform Engineering for Enterprise AI
Backend AI, API and infrastructure engineering for an agent platform serving a unicorn startup and Fortune 500 companies.
Case study
2 min read
AI-Driven Supply Chain & Logistics Platform
An AI-native SaaS platform that replaced manual, document-heavy logistics workflows with automated extraction, routing, and optimization.
Article
3 min read
Database Scalability: Beyond Vertical Scaling
Vertical scaling buys time, not a future. A practical look at query optimization, rate limiting, caching, read replicas, and when sharding is actually worth it.
Article
3 min read
What a Modern TMS Actually Buys You
Transportation Management Systems quietly decide a logistics operation's margins. Where the savings come from: routing, rate shopping, and carrier selection.
Article
2 min read
Leading Engineering Teams as a Fractional CTO
Running distributed engineering teams without a full-time seat: the communication, time-zone, and process realities that decide whether the work ships.
Article
2 min read
Modern Serverless APIs on AWS
For the right workload, serverless lets a small team ship a scalable API with almost no infrastructure to run. Where it fits and where it does not.
Guide
The method, demonstrated
Free
Harness engineering: how we build with agents
Inspect a tenant-scoped export: its contract, observed output and verification record.
Guide
Inspection guide
Free
What to inspect in an agent-built codebase
A practical review of behavior, permissions, maintainability and handoff. Free to read.
Article
Newsletter
Substack
Enterprise AI Patterns
Writing on what it takes to ship agent systems: eval harnesses, tool boundaries, and the rollout discipline they need. Read on Substack.
Guide
Field manual
Free
AI Launch Readiness Checklist
Thirty questions across use-case fit, architecture, data and permissions, evals, observability, and ownership. Work top to bottom before you launch.
Read Enterprise AI Patterns
Writing on what it takes to ship agent systems: eval harnesses, tool boundaries, observability, and the operating layer around frontier models.