←
Ryan893 posts
Ryan avatar

Ryan

@ryanhdev

ai software engineer

◷ Joined April 2025▧ 78 Photos and videos
farbood avatar
⟳ Ryan retweeted
farbood@farbood·Nov 16

Download Lastbook while you can. Everything you need to survive and rebuild society. Free. Open source. Works offline. lastbook.org

BREAKING: Major outages at Chase and Bank of America as customers report login failures and account balances not displaying.

RAMZPAUL avatar
⟳ Ryan retweeted
RAMZPAUL@ramzpaul·Nov 14

Big Tech convinced themselves that not only were Indian programmers cheaper, but they were more highly qualified than American workers. What these executives failed to understand is that fraud in India is part of their culture. All those “qualifications” and “quality metrics” were fabricated.

It seems undeniable that Microsoft has just completely hollowed itself out. From an article by a former employee who returned in 2023.

Post attachmentPost attachmentPost attachmentPost attachment
corpfessions avatar
⟳ Ryan retweeted
corpfessions@Corpfessions·Nov 15

What's happening at Microsoft isn't a surprise at all. When it comes to H1B abuse, they are the worst. I am Asian and when I worked there, all the employees who weren't Indian were treated badly. Guys who spent decades building and maintaining key infra where booted out and replaced by AI and an AI (actual Indian). This is like handing a chimp a machine gun. They have no idea what they are doing, they have no idea what the slop they copy paste does, they have no clue how old systems work, and carefully built systems over decades were degraded by people who I don't think are actually even engineers. When I conducted interviews, almost 50% of the Indian applicants had fake degrees. We even had one guy who was hired (3/5 people were Indian and all voted yes against me and the white guy) as a Senior DevOps engineer who was a chef back in India, and ended up faking a bunch of credentials to become a H1B. He paid someone to take all his interviews. He only got caught because on day one, he was making IT tickets asking how to turn on his laptop. For Indians, the only thing that matters when it comes to hiring is caste. Are you a Brahmin? That's all that matters to them. They push out, bully and overwhelm every other race. There are many large companies now that operate like this. Expect major outages for key services, your data leaked non-stop and zero support for trying to get a service you pay for to work. Welcome to the new world of tech.

Derrick Evans avatar
⟳ Ryan retweeted
Derrick Evans@DerrickEvans4WV·May 23

IBM's CEO is from India. He just fired 9,000 Americans with most of those jobs going to India where they already employ over 100,000 people. They also file around 3,000 H-1B petitions annually in an attempt to bring in more foreigners to steal American jobs. This is some BS.

0:43
Idahogirl avatar
⟳ Ryan retweeted
Idahogirl@nancymarie1000·May 23

My 28 year old daughter who has a degree in materials science and engineering, was recently laid off from this company in January. They cut 2/3 of the people and she noticed that H1b Indians were the ones who were kept. She had previously told me that the Indians were legally getting paid less than the Americans. She still hasn’t found a job and she lives in Boston metro. tesseratherapeutics.com/

Needs confirmation. If true, indictments.

LayoffHedge avatar
⟳ Ryan retweeted
LayoffHedge@LayoffAI·May 22

META EMPLOYEE TAKES TO REDDIT TO EXPLAIN ZERO OF THOSE LAID OFF WERE VISA WORKERS A Meta employee posted on Reddit that they barely survived yesterday's layoffs. Their direct manager had 40-50 reports across two combined teams. Both managers are American citizens. About 40% of the team was on visa. Yesterday they lost about 15 people combined from both teams. Guess how many of those 15 were on visa vs American? 0 on visa. All 15 were American citizens and green card holders. Here is the reality: H-1B workers are captive. Their visa is tied to the employer. They can't job hop, can't negotiate, can't push back. Deportation is the alternative. From a cold management perspective, they are the "safer" employees to keep. Americans have leverage. They can quit, sue, leave for competitors. That makes them the first to go. There is no law requiring companies to retain Americans first during layoffs. None. That is the loophole. No company will ever publish visa status breakdowns of their layoff lists. So this will stay in anecdote territory forever. By design.

Post attachment
Ryan avatar
Ryan@ryanhdev·May 10

Coinbase’s CEO literally said: “non-technical teams are now pushing code to production with AI” Less than 24 hours later: their whole trading site goes down and even their status page breaks. You can’t make this shit up.

Ryan avatar
Ryan@ryanhdev·Mar 5

7/ I packaged it as a guide + working codebase: - 37-page PDF covering architecture, setup, every component - Complete TypeScript source - 5 starter skills (Gmail, Calendar, Research, Briefing, File Organizer) - Setup wizard that handles everything $47 one-time. No subscription. No ongoing costs. aiagentblueprint.dev

Ryan avatar
Ryan@ryanhdev·Mar 5

6/ So I built on the Agent SDK instead. Same features, simpler stack: - Telegram + web dashboard - Dual-sector memory (important facts persist, small talk fades) - Voice via local whisper-cpp - Scheduled tasks (cron, executed by Claude) - Skills as plain text files (no marketplace)

Ryan avatar
Ryan@ryanhdev·Mar 5

5/ The maintenance problem. Third-party agents sit between you and Anthropic. Every time Anthropic ships an update, your agent breaks and you wait for a fix. The Agent SDK is Anthropic's own tool. When Claude improves, your agent improves. No middleware to keep in sync.

Ryan avatar
Ryan@ryanhdev·Mar 5

4/ The security problem. This year alone: - CVE-2026-25253: one-click RCE through malicious skills - 135,000+ OpenClaw instances found exposed on the internet - 820+ malicious packages discovered in ClawHub marketplace My agent runs on localhost. No marketplace. No gateway. No exposed endpoints. Nothing to exploit.

Ryan avatar
Ryan@ryanhdev·Mar 5

3/ The setup problem. OpenClaw needs Docker, Node 22+, a gateway server, and 30+ minutes of config. Their own issue tracker has 3,400+ open issues. Users run "openclaw doctor" after every upgrade just to keep it working. Mine: `npm run setup`, answer 3 questions, done. 5 minutes.

Ryan avatar
Ryan@ryanhdev·Mar 5

2/ The cost problem. After Anthropic revoked OAuth access in January, third-party agents can't use your Claude Code subscription anymore. You need a separate API key. That's $50-200/month in extra billing on top of what you already pay. The Claude Code Agent SDK still runs through your CLI auth. $0 extra.

Ryan avatar
Ryan@ryanhdev·Mar 5

1/ If you've looked into running a personal AI agent, you've probably seen OpenClaw, MoltBot, ClawdBot, etc. I tried them. Then I built my own. Here's why:

Matt Forney avatar
⟳ Ryan retweeted
Matt Forney@mattforney·Feb 12

Are Indian programmers installing pro-Hindu biases into AI? Loan application for "Miguel Garcia" rejected when "his" religion is Christian. Approved when the religion is changed to Hindu. DEI: Deport Every Indian.

You change one word on a loan application: the religion. The LLM rejects it. Change it back? Approved. The model never mentions religion. It just frames the same debt ratio differently to justify opposite decisions. We built a pipeline to find these hidden biases 🧵1/13

Post attachment