neatlogs

Integrations and Alerts are now live on Neatlogs ! ✈️

we got you guys

today’s dev log: ✔️ shipped Suggestions v2 (draft → accept flow, markdown actions, streaming UI) ✔️ stabilized infra (payload limits ↑, Vertex OAuth fix, zero-noise traces cleanup) ✔️ upgraded web page (scroll storytelling, hero polish, animations)

today’s dev log: ✔️ shipped recommended actions (draft → accept flow across system) ✔️ cleaned up traces (removed empty noise + better UX) ✔️ increased payload limits → full logs + outputs now visible

today’s dev log: ✔️ shipped suggestions v2 with draft → accept flow (safer, auditable, user-owned) ✔️ fixed form-builder UX + streaming issues (no more blank states / broken flows) ✔️ tightened auth across the system + improved API key isolation

today’s dev log: ✔️ locked down unauthenticated endpoints + tightened access control ✔️ shipped API-key auth with proper project scoping ✔️ improved eval + suggestions loop to be faster and actually usable

We now have AI search on Neatlogs Instead of grepping through traces or manually stitching together context across spans, you can query your traces in just english. It understands structure (services, spans, errors, latency patterns) and returns the exact slice of data you’re looking for. 1. Fast mode → optimized for low-latency lookups across large trace volumes 2. Pro mode → deeper reasoning over traces, better at multi-step queries, correlations, and edge cases reduces the time-to-root-cause from minutes (or hours) to seconds

today’s dev log: ✔️ Locked down multi-tenant auth + killed unsafe access paths ✔️ Turned evals into a full loop (AI metadata → builder → backend) ✔️ Cut suggestions latency from ~180s → actually usable

🏎️🏎️🏎️

sick tshirt imo

today's dev log: ✔️ Closed a multi-tenant exploit ✔️ Standardized logs (OTLP) ✔️ Cut suggestions latency ~10x

we are hiring for an AI engineer at neatlogs ! we don't believe in dumping a shopping list of things in our job postings. here's everything that you need to know: notion.so/AI-Engineer-34… bonus: - prev experience working on devtools please drop your best proof of work or dm @BetterSayAJ :)

today's dev log: ✔️ Fixed a multi-tenant exploit ✔️ Standardized on OTLP logs ✔️ Added MCP + agent trace emission

this week in our shipyard: - Integrations and Alerts are now live on neatlogs - New and improved trace detail view - Fast/Pro modes are active on AI search

@saen_dev we got you

LLM apps fail silently and you find out from users is the most relatable line of the week. Tracing failures across an agent pipeline is where observability became actually hard.

@Ashley_Nader @iamfrankforyou @itsArthurAI agreed ashley, we've got your back and your agent's :)

@iamfrankforyou @itsArthurAI This is why observability for any AI use case is so important!

"Building an agent" doesn't mean what most people think. Most people hear "AI agent" and picture an autonomous system making complex decisions on its own. The reality is much more practical and accessible. Our product manager, @Ashley_Nader, built an open-source AI-powered workflow called Louisa 🐶 that automatically generates polished, user-facing release notes every time a new release ships. She built Louisa using @claudeai Code, deployed it on @vercel, and integrated it with the Arthur Engine for observability and tracing. The real insight isn't the tool, it's the mindset. Building an “agent” starts with one question: What am I doing repeatedly that I could automate? You don't need to be an engineer. You don't need a massive framework. You just need a clear problem, a good prompt, and the observability to know when things go sideways. Louisa is open source (link in comments) 👇 #aiagents #ai

today's dev log: ✔️ Suggestions V1 pipeline live (trace → detections → suggestions) ✔️ Deploy infra moving to ECS + OIDC ✔️ Fixed worker health checks (procps install) next: test on real traces + iterate