Product blog
Product blog
Product updates, field guides, and field notes—each with links to docs and the hub where relevant. High-level milestones also appear on the changelog.
Topic
ReadField guide · Jun 5, 2026
Running a multi-agent research sprint before touching code
How to use parallel agents to understand a problem completely before changing anything, and why that front-loaded research pays for itself on large work.
- Research as its own sprint
- Parallel investigation agents
- Research-first beats patch-first
Engineering
ReadField guide · Jun 5, 2026
Managing the context window on a large refactor: what broke and how we fixed our sessions
A workflow lesson, not model marketing: how to structure a large refactor so it survives the limits of what can be held in working context at once.
- The whole problem won't fit at once
- Structure work into self-contained units
- Durable artifacts over live memory
Engineering
ReadField guide · Jun 5, 2026
Honest AI tiers: Lite, Standard, Pro — sized in gigabytes, not hype
Why naming real model sizes and recommending a tier your device can run beats promising everyone the heaviest model.
- Lite 0 / Standard ~400MB / Pro ~2GB
- recommendTier() probes WebGPU + memory
- Delete downloaded weights any time
NSS Background Remover
ReadField guide · Jun 5, 2026
Etsy product photos that convert
What Etsy buyers respond to, the image specs and square-thumbnail rule, the mix of lifestyle and detail shots that converts, and how to prep it free and on-device.
- The thumbnail earns the click
- Lifestyle + scale + detail mix
- Consistent, free, on-device
NSS Background Remover
ReadField guide · Jun 5, 2026
Dropshipping, honestly: how the model actually works
What dropshipping actually is, where the money goes, the real advantages (low capital, no inventory), the real problems (thin margins, no control, slow shipping), and an honest read on who the model suits and who it does not.
- How it really works
- Where the margin goes
- Who it actually suits
Novus Stream Solutions (hub)
ReadField guide · Jun 5, 2026
Error handling and alerts for no-code flows
How to handle errors in a no-code or low-code flow without drowning in alerts: backoff, dead-letter handling, quiet-but-trustworthy alerting, and knowing when to fall back to a person.
- Retries with backoff, not blind hammering
- Dead-letter handling for the unsavable
- Alerts you will actually read
Novus Stream Solutions (hub)
ReadField guide · Jun 4, 2026
The approver model: running a build pipeline where AI writes and you review
The operating model for one person directing AI execution: what the human keeps, what the agent takes, and how to keep the review meaningful.
- Human as approver, agent as executor
- Where it works and where it breaks
- Review discipline that scales
Engineering
ReadField guide · Jun 4, 2026
Guardrails and human review: where we let the agent run and where we don't
A practical map of where AI agents operate autonomously and where a human checkpoint is non-negotiable, and the reasoning behind each boundary.
- Autonomous zones vs hard gates
- Reversibility decides the line
- The build as a tireless reviewer
Engineering
ReadField guide · Jun 4, 2026
Multi-band beat sync: triggering bass, mid, and treble independently
Why splitting the audio into bands makes a visualizer feel like it is actually listening to the music.
- Independent bass / mid / treble triggers
- Built on the 32-band FFT
- Motion that tracks the arrangement
Novus Visualizers
ReadField guide · Jun 4, 2026
Keep, kill, or shelve: a structural "no" as a product filter
An operating note on using a structural constraint — must run almost for free — as an honest product filter, not a failure of execution.
- Runs-almost-free as a hard filter
- Shelved AI-agent + sports projects
- A structural no is a complete reason
Operating model