Vikas Bhatt (@dbdnvikas) header
Vikas Bhatt (@dbdnvikas) profile picture

Building and Ranking is what I do (used to?). Here since the Panda/Penguin days. Currently exploring AI + SEO.

India
Joined March 2010
Vikas Bhatt retweeted
Our full CI pipeline is now sub 1 min. Here is what I did: - Vitest -> bun test - tsc -> bun check - oxlint -> bun lint - oxfmt -> bun fmt - shards filled by recorded test time - PRs run only the tests the diff can reach - backend builds the app once per process - lint, typecheck, build and tests all in parallel 200+ hours of waiting saved every week
94
37
15
1,702
150,166
Vikas Bhatt retweeted
Your website can rank #1 in Google and still not be the brand AI recommends. I've been talking a lot recently about query fan-out, entity optimisation, and how AI systems retrieve information before generating an answer. But there's another part of the equation. **What does the rest of the internet say about your brand?** Because your own website is only one source of information. I've broken this down into what I call the **Brand Authority Stack for AI Search**. 8 layers: 1. Website: Establish exactly who you are and what you offer. 2. Backlinks: Earn relevant references from credible websites. 3. Mentions & PR: Get your brand discussed in the right industry contexts. 4. Reviews: Build genuine customer feedback across relevant platforms. 5. Reddit: Understand and participate in real community conversations. 6. YouTube: Demonstrate expertise through videos, reviews and tutorials. 7. Experts: Build relationships with credible industry specialists. 8. First-Party Data: Publish original research and evidence others can reference. Here's where I think SEO is evolving. For years, we've focused heavily on what happens on our websites and which websites link to us. But AI search can also surface what customers say, what experts publish, what communities discuss and what independent publications report. And those sources can tell a very different story from your homepage. This connects directly to what I've been saying about DEO. **Getting found is one thing. Getting cited is another. Getting chosen is the real commercial objective.** The question isn't simply whether AI knows your brand exists. It's whether there's enough credible evidence to make your brand worth considering. Which of these eight layers do you think most brands are neglecting? #SEO #AEO #GEO #DEO #BrandAuthority #AISearch
13
1
13
647
Google judges your page before it even crawls it. In my new episode with Search with Sha, I explain how Google uses predictive information retrieval to cut processing cost. I also cover why visual blocks decide how your page is chunked, and how Topical Authority ranks websites across Google Search, AI answers and iGaming SERPs. Topical Authority = Topical Coverage × Historical Data / Cost of Retrieval We also talk about External Topical Maps, canonical confusion, expired GBP domains, core update timing, Query Deserves a Page, source context, entity attribute value triples and why YouTube and Reddit strengthen your web entity. ▶️ youtu.be/JjRqqRKjdkA #SEO #TopicalAuthority
3
9
1
52
6,662
Vikas Bhatt retweeted
New skill: modern-web-guidance Your agent looks up Chrome's Modern Web Guidance, applies it to your code, then checks the result in a real browser Layout shift, reduced motion, dark mode, dialog focus, more Start here: npx skills add vercel-labs/agent-browser
39
103
16
1,819
138,192
Vikas Bhatt retweeted
EmDash 1.5 is here, with collaborative editing. Your team can now write together in real time. Revisions keep every version you save, new sites stay out of search until launch, and moving from WordPress is smoother. Here's what's new 🧵
19
63
16
685
42,371
Vikas Bhatt retweeted
> Be this marketer > Buy a domain for $11.25 (cheaper than lunch) > Spend an hour on a 3-page site for a deodorant that doesn't exist > Write one Substack post about it > Skip the fake reviews and the Reddit spam > Leave it alone for 3 weeks > Ask ChatGPT two specific questions and see your fake brand named first in all 4 answers > Find it listed above a clinically tested brand carrying the National Eczema Association seal > Learn that ChatGPT recommends whatever it can find a clear page about, even a product nobody can buy Your business has years of happy clients behind it. If a fake deodorant got named in 3 weeks, ChatGPT can name you too. It needs one page that says exactly what you do and which city you do it in. Build that page this week, anon.
28
54
6
1,643
130,973
-> Expired domains -> Exact Match Domains -> Private Blog Networks -> Topical Authority Sitemaps -> Parasite SEO -> Sneaky Backlinks -> Niche Edits and Guest Posts Stack as many of these ontop of eachother as you can to succeed in 2027..
8
5
2
89
7,191
1 video. 140,000 views in 3 weeks. It also made our blog page and product page rank higher. Here’s what happened. We turned our topical map into a video sprint. Every blog post and commercial page became a well-defined AI-generated video for YouTube and TikTok. Same topic. Same facts. Same brand. The numbers 1 video passed 100,000 views. The channel passed 140,000 views in 3 weeks. Then the rankings moved. The video topic built Historical Data for the brand. The blog page on the same topic started to rank higher. The blog page lifted the related product page too. 1 topic got stronger on every surface at once. Why it works Consensus When your brand says the same thing on YouTube, TikTok and your website, LLMs and Gemini see proof instead of a single claim. Cost of Retrieval A brand with consistent facts is easier for search engines and LLMs to find, rank and cite. SERP Estate Your brand takes more spots for the same search, from video results to web results. Source Context Your brand’s link to the topic grows beyond the borders of 1 domain. The formula behind it Topical Authority = Topical Coverage × Historical Data / Cost of Retrieval A video sprint expands coverage, builds Historical Data faster and lowers the Cost of Retrieval. Your topical map isn’t only a content plan. It’s a video plan, a social plan and an entity plan. Brand-entity optimization goes beyond website optimization. Save this for your next content sprint. To learn more: seonewsletter.digital/subscr… #SEO #SemanticSEO #TopicalAuthority
3
4
41
2,432
Vikas Bhatt retweeted
use these 3 condom repos to F*CK REA (reverse engineer anything) it's protection for your site, app or game from hackers and reverse-engineering, for free so REA is died. here's how to use and setup it: first understand how it works: • TruffleHog (29K stars) - hackers reverse engineer apps to steal hidden keys. Anthropic exposed one who scanned 1.8M decompiled Android apps with this exact tool, run it first, k*ll your live keys • BLA (3.7K stars) - hits REA itself: fake keys and decoy code make AI burn tokens on dead ends, your users get the exact same app • Cloudflare security-audit-skill (66.7K stars) - reverse engineering is how hackers find holes. This skill makes Claude Code find them first here's how to put them on for free: 1 → open Claude Code in the folder of your app, site or game 2 → say: "install Node.js 24 and TruffleHog if I don't have them, build my project, then scan the code, its git history and the build folder with --results=verified" 3 → every live key it finds: revoke it, make a new one and keep it on your server, anything inside your app, anyone can pull out 4 → say: "install Cloudflare's security-audit skill with npx skills add cloudflare/security-audit-skill security-audit -a claude code, read its SKILL.md and run a security audit on this project" 5 → fix everything it marks confirmed, then run it again. Cloudflare says one run finds only about half of what repeat runs catch 6 → say: "rebuild, then run npx bla-agents setup and npx bla-agents harden" 7 → the step everyone skips: ship the files from the new .bla-hardened folder, not your old build send repo links below ASAP:
REA is most dangerous thing on the internet right now... It lets your AI agent reverse engineer any app, down to the binary If you use it correctly, you can rebuild any feature you pay for, for FREE In this article, I show you how:
44
143
4
1,273
50,746
Vikas Bhatt retweeted
Every time we have a new model, I run the same simple prompt on all my projects: "Read this entire project and find everything that’s broken, inconsistent, inefficient, or no longer needed. Fix what you’re sure about and list the rest for me to review." highly recommend trying
352
539
228
10,089
1,179,145
My EMD internal-linking playbook: 1. The homepage targets the main commercial keyword. 2. Supporting articles target its problems, use cases, questions, comparisons, and long-tail variations. 3. Every supporting article links back to the homepage or main money page from a genuinely relevant paragraph. 4. Closely related articles link to each other, not only upward to the money page. 5. Anchors are descriptive but varied: 5.1 Exact match 5.2 Partial match 5.3 Question-based 5.4 Product or brand name 5.5 Natural contextual anchors 6. I place the main commercial pages and most important content hubs in the header menu. 7. The footer links to priority pages, major topic clusters, and trust pages such as About, Methodology, and Editorial Policy. 8. I don’t place every article in the menu. The goal is to emphasize the pages that organize and support the site, not create a wall of links. 9. Every important page stays within two clicks of the homepage. 10. Each new article links to older relevant content, and I update older articles to link back to the new one. 11. I use GSC queries to discover better anchor-text opportunities after pages begin receiving impressions. 12. I remove orphan pages and avoid creating multiple articles that compete for the same intent Internal linking is the first thing I do when I want to boost a revenue-generating page - add a few more links + better anchor text.
8
5
4
80
6,283
Vikas Bhatt retweeted
Introducing /competitor-strategy Just give your agent a website. You get the best keywords to track, pages to create, and links to go after, straight from its competitors. Try it on your site: seo.party/skills/every-app/c… Setup below 👇
3
6
139
10,343
Vikas Bhatt retweeted
Fixes are available for CVE-2026-109518, a High severity vulnerability in APT, the package manager used in Debian, Ubuntu, and derivative Linux distributions. The CVE has a CVSS 4.0 score of 7.7. Remediation details are available on our blog. ubuntu.com/blog/apt-vulnerab…
67
348
62
1,653
131,197
‼️ Attackers are scanning the whole internet for Linux servers exposed to a critical 9.8 kernel flaw, Marcus Hutchins says. They found 2.1M servers that accept it. Ubuntu still has no patch. The bug, CVE-2026-98432, is a race condition in the code that handles incoming IPv4 TCP connections. It can let a remote attacker run commands as root during the handshake.
89
519
79
4,209
275,710
Vikas Bhatt retweeted
My 5 month old directory site hit 3k clicks the last 28 days.
13
3
88
4,308
Vikas Bhatt retweeted
Very interesting SEOs: Google says to ignore tactics like chunking content ...but a Google technical writer crafted a guide on how to write "chunk-friendly docs". Advice includes: - Make specific headings - Keep sections under 300 words of prose - Use actual headings - Limit using H4, H5, H6 tags - Avoid long tables
14
6
4
106
5,649
Once I cover the entire content map of a small affiliate EMD, usually with no more than 100 articles, I stop publishing and do this: I connect GSC with Codex and optimize the existing content by: 1. Adding queries that appear in GSC but aren’t mentioned on the page 2. Turning recurring questions into new H2s and H3s 3. Adding direct answers to long-tail queries 4. Expanding sections that don’t fully satisfy the search intent 5. Updating outdated prices, offers, specifications and facts 6. Adding new data and supporting sources 7. Creating comparison tables and clearer decision rules 8. Adding my own experience when I’ve tested the product 9. Building new internal links to and from the page 10. Rewriting internal-link anchors using the language searchers use 11. Improving titles, introductions and headings for stronger relevance 12. Merging overlapping pages that compete for the same queries My websites are intentionally small and highly niched. From there, my growth loop becomes: Off-site: Branded profiles + parasite pages + mentions, links and authority I publish less and improve and distribute more.
8
2
53
2,246
Vikas Bhatt retweeted
卧槽,刚刚检查了一下几台 VPS,有一台机器 30 天被爆破 56 万次,另一台 35 万次,还有一台 19 万次 这周的 Linux 内核漏洞 CVE-2026-98432:不用账号、不用你点任何东西,往 VPS 发几个 TCP 包,就有可能拿下整台机器的权限 CVSS 9.8,据称利用代码是 AI 模型先写出来的。 我自己的 5 台 VPS 全在受影响范围里,刚刚逐台处理: · Debian 已有官方补丁,确认跑在修复版内核上 · Ubuntu 和 CentOS 官方补丁还没正式发,先上 nftables 规则,丢掉带 IP options 的 TCP 包,重启后自动生效 · 顺手做了入侵排查,暂时没发现被入侵的痕迹 所以 @levelsio 说得对:VPS 的公网端口能关就全关。管理走 Tailscale,对外服务走 Cloudflare Tunnel 自连 Cloudflare SSH 都关掉吧,现在就处理! 修复办法,直接将下面这段话 发给大模型,让他去处理: 帮我检查并修复服务器上的 CVE-2026-98432(Linux 内核 TCP 远程漏洞)。 服务器:[填写自己的服务器,我是说我Tailscale 网络里面所有的 VPS 服务器] 1. 先查官方修复状态。Ubuntu 用 pro fix --dry-run CVE-2026-98432,Debian 看 security tracker,RHEL 系看 Red Hat RHSB-2026-016。-proposed 或 testing 仓库的内核不算修复。 2. 逐台检查运行中的内核,区分「补丁没装」和「装了还没重启」。 3. 有官方补丁:只更新内核,保留旧内核,重启前先问我。 4. 没有补丁:用 nftables 丢掉带 IP options 的 IPv4 TCP 包(ip protocol tcp ip hdrlength > 5 drop),并让它开机自动加载。这只算缓解。 5. 验证内核、SSH 和原有服务都正常。不要运行漏洞利用代码。 6. 每台都做只读的入侵排查:成功登录的 IP 和密钥、暴力猜密码的次数、陌生账号和密钥、cron 和 systemd 持久化项、可疑进程、被改过的系统文件、内核崩溃记录。排查过程不改任何东西。 7. 改 SSH 和端口前先问我。目标:管理走 Tailscale,对外服务走 Cloudflare Tunnel,其余端口全关。先确认 Tailscale 能登录,再关端口。 8. 用表格汇报每台的结果。「没发现入侵痕迹」不等于「没被入侵」,要写明排查的范围和限制。 不要让我贴密码或私钥。
This is why I said you want your VPS ports all closed Then only let it be accessible inside your Tailscale for SSH and via a Cloudflare Tunnel to receive web traffic Any Linux box can be taken over with just a single TCP port exposed to the internet now A stop gap is disabling MPTCP on your VPS now but even better is not having any public ports exposed!
29
45
5
395
66,031
If you're running your apps on your own VPS, like Hetzner or Digital Ocean, you should give devpu.sh a try. It's free (MIT license), super easy to deploy, and you can then have a Vercel-like experience to deploying your Docker Compose apps on your own servers. It even comes with an MCP.
13
11
2
167
20,267
THE AI SEO TECHNIQUE TIER LIST 🔎
10
14
272
24,351