Ian Jennings

Agents are getting really really good at programming, try this agent out: https://buff.ly/4cY7vRk I use this agent every single day multiple times a day to build functions with documentation, error handling, and type validation. It's overpowered.

1

Dev agencies, if you don’t want to be treated as a commodity, stop acting like one. - Learn the business context for engineering needs - Stop squeezing your clients every opportunity you get - Prioritize transparency, communication, and accountability In short, position yourself as a partner that uses tech to solve business problems. You really can’t complain about being seen as a commodity if you treat everything as a transaction.

1

1 Comment

SaaS is dead because one can now craft their own backend, not stack it. Major difference. SeaaS (service-as-a-software) is the way to custom your own backend, at last. And the BigFours need to adapt damn fast or they will be reduced to low-value tasks within the next couple of years. This slide is part of the deck that we've been showing during 2024. Most of the ideas will be echoed during 2025 by your standard BigTech as the future. Well, they are not the future for the most advanced agents of the tech industry - they are our present and already our past. SmallTech 1 - BigTech 0 #smalltech1bigtech0 Let's keep it up!

7

2 Comments

TypeScript + Sanity + GROQ just keep getting better together. Now with *automatic* type inference for query results — it's basically magic! 🪄 👀 Keep an eye on the Sanity Changelog for updates like these! https://lnkd.in/eW7eVmtr

76

4 Comments

NEW: Internal Threads Internal threads allow you to spin up conversations in Slack from Pylon. It's great when you want to start a conversation with someone, but they primarily use Slack (e.g. engineers, product people or anyone who isn't living out of the support system). Common internal threads we are creating at Pylon: 1/ #bugs to report a potential bug 2/ #project channels to ask a team a question about a specific feature 3/ #feedback to report product feedback to the team It's been one of the most game-changing features we've released and we're so excited to share it with everyone. Enjoy!

44

3 Comments

Thank you to Jason Lengstorf for talking with me at CityJS Conferences about reducing stress for #softwareengineers, and sharing practical tips for advancing a coding career. + Get a sneak peek at "Web Lunch", Jason's latest series. Link down below!👇

10

1 Comment

Been doing some actual hands-on coding in the last week plus, and I can't say enough about Claude Dev. It would probably annoy expert coders (too slow and asks for too many approvals) and bewilder non-coders (too fast, makes sweeping changes that you have to approve) but for someone like me who knows what code looks like, wants to understand what the bots are doing (and sometimes redirect them) and can frame good questions about error messages, it is incredible. I've been using Cursor in parallel (including Composer mode) to do bits and pieces of this task, and Claude Dev is really a cut above. And right when I thought it couldn't get better, cached mode showed up and saved me 90% of the API costs lol Check it out, well done Saoud Rizwan: https://lnkd.in/e3NiYkes

5

Rails 7.2 now makes sure jobs get enqueued after the current database transaction is committed. Previously, your jobs would run before the records were saved to the database causing errors, so Rails takes care of it for you. If you've ever had to remind yourself (or your team) to use `after_create_commit` callbacks instead of `after_create`, this solves that problem. ActiveRecord uses transactions behind the scenes, so it's an easy mistake to make. And with jobs being retried, it's confusing why the job fails then succeeds a second later. This is another great example of conceptual compression in Rails! https://lnkd.in/gyNtiPnr #ruby #rubyonrails #rails

141

9 Comments

In a vanilla RAG application, you normally chunk the document through either of the following two approaches: 1. Even chunks based on characters. 2. Even chunks based on tokens. 3. Keep a chunking threshold and an overlap parameter. A more nuanced chunking strategy for at least an MVP (not production-grade RAG), would be something as follows: 1. Preprocess the document into elements such as titles, subtitles, headers, footers, page numbers, section headings, table captions, etc. 2. Save the metadata with the elements. 3. Apply chunking conditions based on metadata. 4. Use a chunk-combining strategy if the chunking threshold is not reached. 5. Now vectorize the chunks. #generativeai #RAG

8

6 Comments

Here's a quick look at a new PDF + Screenshot gem I've been building. It uses Ferrum to talk to headless Chrome without any other dependencies and integrates nicely into Rails. https://lnkd.in/g65Uwx3k

250

13 Comments

Announcing 𝘀𝗮𝘃𝘃𝘆 𝘀𝘆𝗻𝗰 - keep your late-night breakthroughs and hard-won knowledge readily accessible on your local machine. No network, No problem. Run 𝘀𝗮𝘃𝘃𝘆 𝗿𝘂𝗻 --𝗹𝗼𝗰𝗮𝗹 and get unblocked instantly. With 𝘀𝗮𝘃𝘃𝘆 𝘀𝘆𝗻𝗰, your knowledge is always with you. #ProductivityTools #DeveloperTools #TechTools #LocalFirst

24

Copilot and development: What activities are easy to automate? Hear what Marcel Chabot - Development Manager at The TM Group, Inc. uses Github Copilot for: - Writing up to 50 lines of code - Clean up long and complex code - Write documentation - Explain what the code does How have you been using Github Copilot to make your job easier? Want to see the full 34min interview? Click the link in the comment section. Dynamics Compass #github #copilot #microsoftcopilot #msdynamics #microsoftdynamics #msdyn365 #d365

2

1 Comment

Automation folks: what's your favorite esignature tool? Specifically looking at options with a strong API/webhooks for status changes to document, custom field mapping, and reasonable pricing. Used to be a fan of PandaDoc, but they're moving all of the API to an "optional" feature of Enterprise (which doesn't work for most of our clients).

19

44 Comments

You've been there - stuck in some app, desperately clicking that “HELP” button, only to be met with either a terrible search box or a frustrating AI-powered chat bubble that offers little to no help. The truth is, users getting lost or confused is a big reason why they disengage. That’s why we’ve upgraded Appcues Launchpads to deliver help exactly when and where your users need it. Noticeable and (actually) useful without being annoying. Whether it’s a tour, a link, or a help doc, Launchpads offers self-support that’s perfectly timed, perfectly themed, and perfectly helpful. No engineering required. No annoying “ping” sounds. Just a smoother experience for your users. ps: I'm really proud of the team that brought this to life, specifically Joseph Muething, Alexandre S. and Magdiel Duarte. (But don't poach them. I need them.)

12