←
Lian Lim | Dashboard & AI Automation Expert9,497 posts
Lian Lim | Dashboard & AI Automation Expert avatar

Lian Lim | Dashboard & AI Automation Expert

@dashboardlim

Building 8-9 figure AI Operating Systems

◷ Joined February 2020▧ 991 Photos and videos
Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 3

Claude built your SaaS login? cool. now try breaking it with 2 browser windows. this is one of the simplest tests i'd recommend to anyone vibe coding an app with real users 1. create user A and user B 2. log in as A and open a private record 3. copy the record URL 4. log in as B in another browser 5. paste A's URL B should NOT be able to read or modify A's record now repeat the test against the actual API endpoint because hiding a button in the frontend is not authorization and just because your app has login doesn't mean your users' data is isolated this catches a class of bugs that a beautiful UI will never reveal before you add another AI feature, make sure customer B can't see customer A.

Lian Lim | Dashboard & AI Automation Expert avatar
⟳ Lian Lim | Dashboard & AI Automation Expert retweeted
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 15

one of the most useful dashboards we've built ends with a date not a chart, it's something simple-looking an actual date telling the business when it needs to order stock the client was managing inventory across multiple regions sales numbers in one place stock levels somewhere else forecasting in spreadsheets and by the time someone put everything together, the numbers were already getting old so we connected the data and built a weekly forecasting system now the team can enter a revenue target and see: → how many units need to sell each week → whether current stock can support that target → which products are running low → when the next order needs to be placed the calculation also accounts for a 25-day supplier lead time the system doesn't place orders for them it gives the team the information to make the decision i love builds like this because the best output wasn't another dashboard metric it was knowing when to pick up the phone and order more stock.

Post attachment
Lian Lim | Dashboard & AI Automation Expert avatar
⟳ Lian Lim | Dashboard & AI Automation Expert retweeted
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 16

if your favourite coding model disappears from GitHub Copilot today, it might not be a bug GitHub scheduled 3 model retirements for November 16 → Claude Opus 5 → GPT-5.6 Terra → Claude Opus 4.8 (with an exception for eligible annual individual plans) the suggested replacements include Opus 5.5 and GPT-6.1 Sol worth checking if you have workflows that depend on a specific model especially if you're running: coding agents automated PR reviews reusable agent configurations team-wide Copilot setups model availability changes faster than most engineering teams update their internal documentation and sometimes the model your team has been benchmarking against isn't even the one everyone's using anymore before blaming the agent for behaving differently, check which model actually ran.

Lian Lim | Dashboard & AI Automation Expert avatar
⟳ Lian Lim | Dashboard & AI Automation Expert retweeted
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 16

Vercel just added an interesting way to make AI decisions using live web evidence it's called Parallel Decisions and the part that caught my attention is the output instead of just generating an answer, it can return a probability + cited evidence imagine building an AI lead qualification system you want to know: → does this company operate in the US? → does it offer a specific service? → does it meet a publicly verifiable requirement? you can ask a structured question and get an evidence-backed result not just an LLM confidently guessing from old training data i'd still verify the sources and use human review for important decisions but this is a useful building block for agents that need to act on information that changes every day it's available through Vercel AI Gateway.

Lian Lim | Dashboard & AI Automation Expert avatar
⟳ Lian Lim | Dashboard & AI Automation Expert retweeted
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·19h

here's an AI automation i'd recommend building if you're looking for your first paying client a missed-lead recovery system find a business getting enquiries through its website, WhatsApp or Facebook ads ask them to review their last 100 leads - how many never received a reply? - how many waited hours? - how many never booked? now build a simple workflow: 1. new lead enters the CRM 2. notify the assigned salesperson 3. escalate if nobody responds within 15 minutes 4. draft a follow-up for approval 5. track which leads eventually book start with a small paid pilot show the business how many enquiries were going unanswered before and after you now have something measurable to sell instead of another AI chatbot demo.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 15

try typing a single % into your app's customer search bar seriously, just this: % if every customer suddenly appears, there's something worth checking in your search logic in PostgreSQL, % isn't an ordinary character inside a LIKE pattern it's a wildcard that matches any sequence of characters and _ matches any single character so if your app builds a search using: ILIKE '%search_term%' and the user types % you might end up matching almost everything here's what i'd test: 1. search for % 2. search for _ 3. search for a normal customer name 4. compare the results if the search is supposed to treat user input literally, those special characters need to be escaped before building the pattern and yes, you should still use parameterized queries because parameterization protects against SQL injection it doesn't automatically turn LIKE wildcards into literal text tiny difference surprisingly easy to miss when AI generates your database queries.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 13

GitHub just published a guide on making agent skills work across Claude Code, Cursor and Copilot the part i liked wasn't even the portability it was how simple the whole idea is a folder a SKILL.md file a repeatable task clear instructions on what the agent should do instead of explaining the same engineering workflow in 20 different chats, you can turn it into something your whole team can review and reuse imagine a skill for investigating production bugs it knows: → which logs to inspect → how to trace the failing request → what evidence to collect → what tests to run → what the final report should contain the skill doesn't magically grant tools or permissions those still come from the environment running it but the useful part of your workflow no longer has to disappear inside someone's chat history i'd rather have 10 well-written reusable skills than 100 prompts nobody can find again.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 12

Claude built your ecom dashboard? cool. now refund a $100 order. then check what your revenue chart says this is one of the easiest tests i'd run before trusting an AI-built sales dashboard 1. create a test order for $100 2. mark the payment successful 3. issue a full $100 refund 4. refresh your dashboard 5. check gross sales, refunds and net sales for a simple example excluding taxes and fees, you should see: gross sales: $100 refunds: $100 net sales: $0 if net sales still shows $100, your dashboard might only be counting successful payments now repeat with: → a $25 partial refund → a cancelled order → a failed payment → a payment that gets refunded the next day because your dashboard needs to know the difference between money collected and money kept AI can generate a beautiful revenue chart in minutes that doesn't mean it knows how your business defines revenue.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 11

Claude Fable 5.2 just helped build a working operating system in 8 days. with over 1,000 AI agents that's a pretty insane demonstration but before switching your entire Claude Code workflow to Fable, i'd run one test give Fable and your current coding model the exact same bug 1. start from the same Git commit 2. give both the same task + instructions 3. use the same tools and time budget 4. let each investigate, fix and test 5. compare the final results don't just measure which agent finished first check: → did it actually fix the bug? → did it break anything else? → how many unrelated files changed? → how many times did you intervene? → what did the task cost? a model that finishes in 3 minutes but introduces 2 regressions isn't necessarily better than one that takes 10 and a benchmark score won't tell you how well a model handles YOUR codebase before paying for the smartest coding model, find out if it actually makes your engineering better.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 10

here's a 2-minute test for anyone who built a dashboard with AI. put 101 records in your test database. make sure the API returns 100 per page. now open your dashboard. what does the total say? 100 or 101? i've seen how easy it is for dashboards to calculate totals from whatever the frontend happened to load. the chart looks correct. the API returns 200. there's no error. but the dashboard is missing page 2. to check yours: 1. compare the dashboard total with a database COUNT 2. add a filter that matches records on multiple pages 3. check that charts include all matching records the fix is often to calculate aggregates server-side instead of summing the currently loaded table rows. pagination should change what you display. not what the business thinks it has.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 9

if Claude wants to rename a database column and the app is already live please don't let it happen in one deploy imagine changing: customer_name → full_name the new backend expects full_name but your old backend workers are still running your scheduled jobs still use customer_name and users have requests in flight suddenly a tiny schema change becomes an outage the safer pattern is expand → migrate → contract → add the new column without removing the old one → make the application compatible with both → backfill existing data → move reads + writes to the new field → verify the old field is no longer being used → remove it in a later deployment you need to plan for mixed application versions and concurrent writes during the transition AI can generate the migration in seconds it can't make your existing production traffic disappear while you deploy it never see your database migration as just a code change because it's more of an actual rollout.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 8

before Claude Code touches an unfamiliar repo, i want it to answer 5 questions not tell me it understands everything just this: 1. which file actually owns the behavior we're changing? 2. where does the data come from? 3. what else depends on this? 4. what's the smallest possible fix? 5. how will you prove it worked? you can literally paste: “investigate first. answer these 5 questions with file references. propose the smallest change. don't modify anything during this investigation.” it's amazing how much bad code you avoid by making the agent understand the problem before it tries to solve it the first useful output isn't always code I feel sometimes it's a map of what shouldn't be touched

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 7

here's something i'd make Claude Code do before fixing a bug that involves both frontend + backend trace the entire request. not just the file where the error appeared paste this: "Investigate this bug without changing code. Trace the full path: 1. user action in the UI 2. frontend handler 3. API request + payload 4. backend route 5. database read/write 6. API response 7. UI state update For every step, show the exact file + function. Mark anything you couldn't verify as UNKNOWN. Identify the earliest point where expected and actual behavior diverge. Do not propose a fix until the trace is complete." this forces the agent to distinguish code it actually inspected from code it assumes exists especially useful when the frontend says one thing and the database says another sometimes the fastest way to fix a bug is to stop the agent from fixing it for five minutes.

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 6

we spent time building an automation tested it watched it run then switched it off a few days later not because it broke it did exactly what we'd built it to do the problem was what it replaced a monthly reporting process that forced us to sit down, look at the numbers and actually think about what was happening we'd automated the report but we'd also removed the moment where we made decisions so we killed the automation there's something funny about running an AI automation company and realizing that sometimes the manual step is the valuable part what's one task in your business you'd deliberately keep human?

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 5

this is more interesting than another AI model getting smarter Claude can now hand a plugin a long-running job and keep working the plugin can also stop mid-task and ask a human for something it needs sounds small until you've built a real automation exports take time APIs fail humans disappear halfway through approvals quite sure the problem isn't just making agents autonomous it's deciding what happens when a task is halfway done and nobody answers for everyone building MCP tools... do you make the task wait, fail, or continue with a fallback?

Lian Lim | Dashboard & AI Automation Expert avatar
Lian Lim | Dashboard & AI Automation Expert@dashboardlim·Nov 4

we built 255 rules so our AI could ignore 5,536 emails. yes, ignore them. one of our clients had 7 mailboxes full of newsletters, notifications, customer enquiries and random noise the obvious solution was to let AI read everything and decide what mattered we did the opposite we put a boring filter in front of the model blocklists allowlists pattern matching rules the client could edit themselves over one 30-day period: → 5,536 messages filtered before AI → 2,256 processed downstream → 1,052 replies drafted and every filtered message still had a recorded reason the funny part is that we spent so much time building an intelligent inbox... only to discover that one of its best features was knowing when not to use intelligence at all sometimes the best AI optimization is a simple rule.