Activity
Mon
Wed
Fri
Sun
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
What is this?
Less
More
AI Phone Automation
24
Free
Vibe To Exit
130
Free
Meta Muse For Work
353
Free
Rapid Flow Automation
39
Free
Clyde | Open Source AI Agents
114
Free
AI Staffroom
887
Free
Bestapps Academy
260
Free
AI Agents for Experts
37
Free
11 contributions to Vibe Coders
Nov '25 •
What is the biggest bottleneck with current vibe coding workflows
Hey folks, quick pulse check. What’s the biggest bottleneck slowing down your current VIBE coding workflow? Is it tooling, context-switching, messy repos, slow reviews — or something even wilder? I’m looking for real blockers, not the polite ones we tell our managers. Drop the pain points, the annoyances, the “why is this still a thing?” moments. If it kills your vibe, breaks your flow, or drains your sanity — I want to hear it. Let’s surface the friction so we can fix it fast.
0 likes • Aug 26
Integrations are slow because the agent usually lacks the provider’s exact auth scopes and request contract. I use a one-page checklist for each service: auth type, minimum scopes, API version and base URL, required environment-variable names, a sandbox test, and webhook-signature handling; for debugging, give the agent a redacted request/response or OpenAPI schema rather than a secret. The gotcha is separating 401, 403, and 429 failures, so log the status, response body, and provider request ID in a local reproducible test.
Jan 19 •
Recommendations
I am looking for recommendations for an affordable advanced Data Science program. I would prefer an online format, but I am also open to physical campuses within Kenya and abroad. Does anyone have a school or bootcamp they would highly recommend for advanced-level learning?
0 likes • Aug 26
For an advanced program, I’d screen the syllabus for probability and statistics, linear algebra, optimization, model evaluation, and a substantial capstone using messy real-world data—not just notebook exercises. In Kenya or online, compare the full cost and schedule, ask to see recent capstone projects and alumni outcomes, and audit one module before paying; DataCamp can be useful for practice, but it may be lighter on theory than you need.
Jan 29 •
Are you also facing a ban issue - or got your Claude account banned?
People are reporting Anthropic warnings/bans when using Claude OAuth (Pro/Max) through third-party tools like Clawdbot (instead of API access). I'm dropping the few links below - anyone here hit by this, and what workaround are you using (API key / different model / different tool)? https://x.com/elijahsystems/status/2016201425273958453
1 like • Aug 26
The safest workaround is to stop using the Pro/Max OAuth token in the third-party tool, revoke that session, and configure the tool with a separate API key; then test with low request concurrency and no unattended loops. The gotcha is that consumer OAuth and API access have different terms and rate-limit behavior, so a setup that works in the consumer app can still look like automated account use when routed through a wrapper. If you need a fallback, route only the non-sensitive workload to a second model or provider rather than repeatedly retrying the same failing OAuth call.
Dec '25 •
I built an AI that feeds me interview answers in real-time (2 second response time)
Spent the last little bit building this. It listens to the interview through my audio routing, detects when a question is asked, and displays a suggested response on my screen in about 2 seconds. The video shows a basic test - it catches "tell me about yourself" and pulls up a pre-written response tailored to my actual background. But here's where it gets interesting. The next version has what I'm calling the "War Room" - 9 AI agents working together: - Pre-interview dossier builder (researches the company before the call) - Tactical advisors based on Chris Voss and Oren Klaff frameworks - Real-time fact verification so I don't contradict my own resume - Different modes for different job types Basically a team in a van feeding me intel while I'm in the interview. All running locally. No cloud APIs. Just Ollama + Python + a lot of prompt engineering. Would love to hear what other use cases people see for something like this. Sales calls? Customer support? The architecture is modular. https://www.youtube.com/watch?v=Lo8cBitoChU
0 likes • Aug 26
Beyond interviews, the same local Ollama plus Python pipeline fits sales-call objection coaching and support triage, but I’d separate the real-time capture layer from the retrieval and policy layer so latency spikes don’t freeze the UI. For a first production slice, log transcript chunks, retrieve only approved playbooks, and show confidence or source snippets instead of letting the agent invent a response. The main gotcha is consent and retention: make recording visible, encrypt local transcripts, and set an automatic deletion window.
Dec '25 •
Vibe coding tool
Which is best vibe coding ide as per your experience for fixing bugs, performance and security?
0 likes • Aug 26
For a mix of bug fixing, performance, and security, I’d use Cursor or Claude Code with repo-wide context, but keep the workflow explicit: reproduce with a failing test, ask the agent to trace the smallest relevant path, then run the tests and a dependency/security scan before accepting changes. The gotcha is asking one prompt to “optimize and secure everything” — it tends to make broad edits; isolate one concern per branch and inspect the diff.
1-10 of 11
@yousef-halel-2063
Co-Founder at Vendo (YC S26). UC Berkeley honors grad at 19, Stanford MS. Building software that lets companies agentify products vendo.run is the web
Active 9h ago
Joined Aug 17, 2026
Powered by