Blog
Insights & Articles
Thoughts on building great products, engineering culture, and the future of remote teams.
06
All articles
Every article we've published, in one place.
Browse the full archive (76)Collapse archive
2026
- How to Secure and Observe Agent Memory in ProductionTutorials
- Redis and PostgreSQL Together for AI Agents: Cache Fast, Store Truth in PostgresTutorials
- How to Migrate Agent Memory from Redis to PostgreSQL Without Breaking SessionsTutorials
- How to Design a PostgreSQL Schema for Agent Memory and Session StateTutorials
- Law 83/2017 and Real Estate Agents: The Complete AML Compliance GuideIndustry Insights
- AI Governance Is Not a Policy Document: It Is an Operating ModelProduct
- Pinecone vs Modal: Which AI Stack Layer Are You Actually Buying?Industry Insights
- Why NeMo Guardrails Is Better Than Guardrails AI for Safer LLM ApplicationsEngineering
- Why Langfuse Is Better Than OpenLIT for LLM Observability and TracingEngineering
- Why Modal Is Better Than Runpod for Shipping AI Workloads FasterEngineering
- Why Unstructured Is Better Than LlamaParse for Document Ingestion at ScaleEngineering
- Why pgvector Is Better Than Pinecone for B2B Product RetrievalEngineering
- Why Qdrant Is Better Than Chroma for Production RAG InfrastructureEngineering
- Why Portkey Is Better Than LiteLLM for Multi-Model AI Gateway ControlEngineering
- Why LlamaIndex Is Better Than LangChain for Production RAG SystemsEngineering
- Why Triton Is Better Than vLLM for Multi-Model GPU Inference PlatformsEngineering
- Why Direct Model Provider Integrations Are Better Than OpenRouter for B2B AI ProductsEngineering
- Why LangSmith Is Better Than Opik for LLM Evaluation WorkflowsEngineering
- Why Cohere Rerank Is Better Than Voyage AI for Retrieval QualityEngineering
- Why PostgreSQL Is Better Than Redis for Agent Memory and Session StateEngineering
- Why Qdrant Is Better Than Milvus for High-Scale Vector InfrastructureEngineering
- Why Temporal Is Better Than LangGraph for Long-Running AI WorkflowsEngineering
- Why FastAPI Is Better Than BentoML for Production LLM APIsEngineering
- Why Qdrant Is Better Than Weaviate for Production Vector SearchEngineering
- Why Haystack Is Better Than LlamaIndex for Enterprise Retrieval PipelinesEngineering
- Why LangGraph Is Better Than the OpenAI Agents SDK for Production AgentsEngineering
- Why PydanticAI Is Better Than DSPy for Structured LLM DevelopmentEngineering
- Why vLLM Is Better Than Hugging Face TGI for Self-Hosted LLM InferenceEngineering
- Why LangGraph Is Better Than CrewAI for Production Agent WorkflowsEngineering
- Decision Documentation for Distributed Teams: Why Remote Speed Depends on Better Written Trade-OffsEngineering
- Structured Engineering Interviews: A Better Hiring System for Teams That Need Senior Judgment FastCulture
- Legacy Modernization Without a Big-Bang Rewrite: The Operating Model Most Teams Actually NeedEngineering
- MVP Rescue vs Full Rewrite: How to Decide When the Product Exists but the Delivery System Is BrokenProduct
- B2B Product Discovery Without Roadmap Theater: How to Validate the Risk Before You Fund the BuildProduct
- Technical Due Diligence for Software Buyers: The Questions That Surface Risk Before the Contract ClosesIndustry Insights
- Kubernetes Operations Maturity: When the Platform Starts Costing More Than It SavesEngineering
- An Observability Strategy With OpenTelemetry: What Growing Product Teams Should Standardise EarlyEngineering
- Production RAG Systems With Governance: Why Retrieval Quality Is Only Half the JobEngineering
- AI Governance Operating Models: What Separates Responsible Ambition From Expensive ChaosProduct
- Threat Modeling Before Sprint Commitment: A Simpler Way to Avoid Expensive Security ReworkEngineering
- Incident Response Tabletop Exercises for Software Companies: The Fastest Way to Find Process FictionTutorials
- Building an OWASP SAMM Roadmap: How to Improve Security Maturity Without Stalling DeliveryTutorials
- OWASP ASVS as a Release Standard: A Better Way to Define Application Security ExpectationsTutorials
- Secure SDLC With NIST SSDF: What Buyers Should Expect From a Product Engineering PartnerEngineering
- NIST CSF 2.0 Gap Assessment: Turning a Framework Review Into an Operating PlanTutorials
- How the Cyber Resilience Act Should Change a Product Roadmap Before Certification Questions ArriveIndustry Insights
- NIS2 Gap Analysis for Growth-Stage Companies: What to Fix First When Everything Looks PartialIndustry Insights
- DORA Compliance for Software Vendors: Why Financial Buyers Now Audit Delivery Quality TooIndustry Insights
- AI Act Compliance for Product Teams: What Needs to Change Before the Deadline Changes YouIndustry Insights
- Dedicated Teams vs Staff Augmentation: Which Model Creates More Value for Your Product?Industry Insights
- What an AI Implementation Strategy Delivery Plan Needs to Include to Move Beyond PilotsProduct
- How to Choose a Software Development Partner Without Buying More RiskProduct
- Prototype Development Services: What a Prototype Should Actually Prove Before You BuildProduct
- How Cybersecurity Audits Reduce Risk When They Go Beyond the ChecklistIndustry Insights
- 7 NIS2 Readiness Mistakes That Put Compliance and Resilience at RiskIndustry Insights
- Building Production AI Systems: What Separates a Demo From a Reliable ProductEngineering
- What a Security Remediation Program Should Look Like After the Findings Come InTutorials
- What Product Engineering Services Should Deliver Beyond CodeEngineering
- When to Bring in Cybersecurity Advisory Services Instead of Waiting for a CrisisIndustry Insights
- QA Is a Design Function: How High-Trust Teams Build Fewer SurprisesEngineering
- Testing Is a Design Function for High-Trust TeamsEngineering
- QA Is a Design Function: How High-Trust Teams Build Fewer SurprisesEngineering
- Remote Engineering Onboarding Is a Delivery System, Not a Welcome PacketEngineering
- Product Discovery Works Better When Engineering Shows Up Before the TicketsProduct
- Roadmaps That Survive Engineering RealityProduct
- The Distance Between an AI Demo and a Production System Is Mostly PlumbingEngineering
- Cybersecurity Reviews Should Happen Inside Delivery, Not After ItIndustry Insights
- Team Augmentation That Actually Works: Operating Models, Not Extra HandsCulture
- Architecture Decisions Don’t Belong in Hallway Conversations AnymoreEngineering
- Hire for Fundamentals, Not a Shopping List of Tool KeywordsCulture
- Before You Rewrite, Measure the Friction in Your Deployment PathCulture
- How to Actually Build an AI Agent (Without the Hype)Engineering
- ChatGPT 5.5 vs Claude Opus 4.7: Which Tool Should You Actually Use?Industry Insights
- How We're Using Claude's New "Effort Dial" for HR TasksEngineering
- Why Yann LeCun Bet $1 Billion That LLMs Are a Dead EndIndustry Insights
- Building the futureIndustry Insights
05
Build with Alongside
Explore how we deliver product engineering from Portugal — then see it in client work.