Blog
Things we've shipped — and how.
Practical AI engineering. Architecture, guardrails, eval loops, and the production gotchas no one writes about. Filter by tag below.
- 8 min readevals · agents
Evals as a permanent system, not a launch checklist
Why eval suites that ship at launch and then atrophy are the single biggest cause of AI quality decay. The pattern we build instead.
Read → - 8 min readecommerce · agents
The e-commerce AI playbook for 2026
Catalog assistant, support deflection, product copy at scale, review intelligence — what to ship first, in what order, and what each costs.
Read → - 9 min readhiring · team
Hiring for production AI — what to actually screen for
Resume signals that don't matter, interview questions that do, and how to tell a senior AI engineer from someone who can run a notebook.
Read → - 9 min readmanufacturing · agents
Manufacturing AI readiness — the 6 questions that decide v1
Multi-agent ops, predictive maintenance, document automation, AI-native ERP. Which one you build first depends on six honest answers about your plant.
Read → - 10 min readfinance · compliance
AI compliance for fintech: what auditors actually look for
Audit trails, model approval workflows, refusal patterns, and the difference between SOC 2 readiness and SOC 2 audit. The compliance pattern we ship for regulated finance clients.
Read → - 10 min readmanufacturing · agents
We built an AI agent for manufacturing: here's what it cost and what it does
Real numbers — total build cost, monthly run cost, time-to-value, and what we'd do differently. From our Multi-Agent Manufacturing case study.
Read → - 8 min readstrategy · roadmap
AI implementation roadmap: a 90-day plan for mid-market companies
Week-by-week, what to ship and what to defer. The plan we actually run at the start of strategy engagements.
Read → - 9 min readagents · saas
How to build a customer-support AI agent for SaaS
Architecture, guardrails, eval loop, and the deflection math that justifies it. The playbook we use for B2B SaaS clients.
Read → - 7 min readrag · fine-tuning
RAG vs fine-tuning: which one your e-commerce business actually needs
When to use retrieval, when to fine-tune, and when to do both. With a flowchart your team can actually use.
Read → - 11 min readautomation · n8n
n8n + GPT: 7 workflows automating $50K/year of manual ops
Real workflows we've shipped — invoice triage, lead enrichment, content distribution, and four more. With the cost-benefit math.
Read →
