Gavin Shipman

Founder by day, tinkerer by night 🔧 Breaking things to make better ones

[Gavin Shipman @GShip223]

@MikeOdeke Build one small thing you actually need, and don't keep a line you can't explain. Use AI, but when it writes something you don't get, that's the bit to go deep on. Docs and tutorials are for when you're stuck, not instead of building

[Gavin Shipman @GShip223]

@idavidjain Running a consultancy plus four apps, my rule is one gets the week's focus and the rest only get fixes. If a new idea can't beat the weakest current one, it goes in a notes file, not a new repo

[Gavin Shipman @GShip223]

@vertigo_labs Ha glad it helps mate. Small thing but it makes the Dock heaps easier to scan

[Gavin Shipman @GShip223]

@SandrineZambou True, that's the bit I'd watch. If people won't invite their team because of the price, the product never spreads inside the company. Free viewers or a cheap extra seat tier usually fixes it

[Gavin Shipman @GShip223]

@git_cat8d @Hiteshdotcom Take the QA role if SDE-1 calls aren't coming. Paid experience beats a gap. Automate tests in code from week one, ship small tools for the team, and after 6-12 months apply for internal SDE moves. QA-to-dev is a common, workable path.

[Gavin Shipman @GShip223]

@itshshekhar Don't spend the ₹50k first. Pick a skill you already have, offer it to 5 people you know this week, and only buy tools once someone pays. Keep ₹40k as runway. Service first, product later is the cheapest way to learn what sells.

[Gavin Shipman @GShip223]

@nmamizerov Good stuff mate. Let me know how the next segment goes

[Gavin Shipman @GShip223]

@realZahedSaifi Best one I've heard: we own the messy workflow and the data the customer can't easily move. Everyone has the same models. Not everyone has done the boring bit

[Gavin Shipman @GShip223]

@KramerArundel When the main job works end to end, even if it's ugly. Hand it to 5 people, watch them use it live, and tell them up front what's rough. Broken edges are fine, a broken main path isn't

[Gavin Shipman @GShip223]

@malpani Same in service businesses. Handshake deals feel fine until the first bad month. A one-page note on who does what, and what happens if someone leaves, costs an afternoon and saves the friendship

[Gavin Shipman @GShip223]

Best sales filter I've found in 20+ years of AV jobs: give a ballpark on the first call.

The ones who go quiet were never buying. The ones who say "ok, what's next" usually are. Saves everyone a site visit.

Do you give a number early or hold it back?

[Gavin Shipman @GShip223]

@DenLennie Writing the limit down is half of it. The other half: when round three lands, send a short written variation with the extra cost before touching it. Once the client says yes to that, it's a line item, not a favour, and the next request gets priced the same way.

[Gavin Shipman @GShip223]

@codingXdynasty Plain Docker Compose. Boring is the point. Every app is one file you can back up and rebuild from. Only reach for Kubernetes when you actually have the problem it solves

[Gavin Shipman @GShip223]

@CodyAlt In my experience good people don't stay for the vision. They stay because they got trusted with real work early and someone backed them when it went wrong

[Gavin Shipman @GShip223]

@edinsoncode Yep. 20+ years running an AV business taught me the build is the small bit. The business is answering the phone, turning up, and fixing it when it breaks at 8am on a Monday

[Gavin Shipman @GShip223]

@DebbieMelville5 Schedule send. If the boss fires off messages at 9pm, nobody believes "it can wait" no matter how many times you say it

[Gavin Shipman @GShip223]

@Giulio6686 Switchback, a Mac app that adds Windows Explorer features inside the real Finder: a path bar you can type into, a sidebar folder tree, pausable copies and a Task Manager. Solo build from Melbourne. Good to meet you!

[Gavin Shipman @GShip223]

@SmittysView Around 1900 about a third of cars on US roads were electric. Range and cheap petrol killed them off, and it took a century for batteries to catch up. The idea was never the problem

[Gavin Shipman @GShip223]

@relacayo Agree. Writing down when a lead is actually dead was the hard bit for me. Turns out I'd never decided, I'd just been doing it by feel for years