About
At Transurban, my focus as a Cyber Threat and Vulnerability Specialist is on safeguarding…
Services
Articles by Hardik
-
Indians and Digital India's Start-up Economy
Indians and Digital India's Start-up Economy
ChatGPT's perspective on this topic: India has emerged as one of the fastest-growing startup ecosystems in the world…
1 Comment -
Indians and Cricket Passion
Indians and Cricket Passion
Cricket in the eye of ChatGPT: Cricket is a hugely popular sport in India, and many Indians are passionate about the…
Activity
4K followers
-
Hardik Cholera shared thisA most comprehensive set of tips I've come across on getting most out of your Claude subscription. How to Claude the right way!! #ClaudeAIComputing #SonnetOpusFableHardik Cholera shared thisYou pay $20/month for a Claude subscription. But you still hit the limit by 2 pm. I did too, until I understood how Claude calculates credits: It re-reads the entire chat every time you send a prompt. Prompt 30 costs about 31x more than message 1. Upload the same PDF to 5 chats & you pay for it 5 times. Run Opus on a grammar check & you pay Opus prices for a Haiku job. Here are the stupid simple 9 fixes to use every day: FIX 1. Shrink your files. Open a Google Doc. Paste the text from your PDF. Download it as a .md file and upload that instead. A 15-page PDF costs ~45,000 tokens. The same text as .md costs ~2,000. FIX 2. Plan in Chat, then build in Cowork. In Claude Chat, prompt: "Help me plan a [financial model]. Ask me questions first." Paste the finished plan into Cowork & say: "Build exactly this." FIX 3. Let Claude prompt you instead. A 500-word prompt costs 500 tokens. A click costs almost zero. So prompt: "I want to [TASK] to [SUCCESS CRITERIA]. Ask me questions using AskUserQuestion before you start." Full guide here: claude.rubenhassid.ai FIX 4. Edit your last message. If you want to correct a mistake, go to your sent prompt, click "Edit" on your message, retype your prompt, and hit Save. Claude regenerates from that point. FIX 5. Summarize every 15 prompts chat. Prompt: "Summarize this entire conversation into a brief." Copy the brief. Paste it as message # 1 in a new chat. Turning 105,000 tokens into 500. FIX 6. Pick the model before you start. Open the model dropdown in the chatbox. Under 30 seconds of thinking? Sonnet. Execution, multi-step work? Opus. Brainstorming or solving a complex problem? Fable FIX 7. Use Projects for repeated files. Create a Project. Upload your files there once. Every chat inside that Project reads them without re-tokenizing. FIX 8. Turn off the settings that a task doesn't need. Go to the tools panel. Turn off Web Search, connectors unless this specific task needs them. Every idle feature adds tokens. FIX 9. Batch your asks. Send "Summarize this, list the key points, and write a headline" as ONE message. Three separate messages make Claude reload the full context three times. Grab the free guides to copy my full system: ✦ The 27-fixes breakdown: claude.rubenhassid.ai ✦ The infographic cheat sheet: how-to-ai.guide
-
Hardik Cholera shared thisHardik Cholera shared thisThe paradox of Indian tech: India has 5.4M software engineers, yet it hasn’t produced a single globally dominant app, OS, database, cloud, or social network. Why? A few reasons: ↳ Services over products - IT outsourcing (Infosys, TCS, Wipro) built billions but not billion-user products. ↳ Brain drain - Indian engineers built Google, Microsoft, Adobe… but in the United States. ↳ Open market - Unlike China’s protectionism, India lets US apps dominate. ↳ Risk aversion - Decades in “survival mode” made safe jobs > bold innovation. On the other hand, India did build something the world envies. Digital public infrastructure. → UPI processes 24B+ monthly transactions (faster & cheaper than Visa). → Aadhaar verified 1.35B people in milliseconds. → ONDC is re-wiring commerce itself. Instead of apps, India built protocols. Instead of unicorns, it built rails. That choice lifted financial inclusion and turned India into the world’s largest real-time payments economy 💸 So maybe India’s not missing an app or AI unicorn. Maybe the world’s missing India’s playbook. P.S. also check out The Age of the One-Person Company (how solo founders are building million-dollar companies with AI)🦄: https://lnkd.in/dFbRmWbx
-
Hardik Cholera reposted thisHardik Cholera reposted thisCVE MCP Server — AI-Powered Vulnerability Intelligence A production-grade MCP server that gives AI agents access to CVE lookup, EPSS, CISA KEV, MITRE ATT&CK, Shodan, VirusTotal, OSV, and more through 28 security tools and 24 data sources. Its triage_cve tool correlates multiple sources, calculates a composite risk score, and provides prioritized vulnerability-triage reports. Source: https://github[.]com/mukul975/cve-mcp-server #cybersecurity #ThreatIntelligence #vulnerabilitymanagement
-
Hardik Cholera reacted on thisHi CIA-A Squad, Highly recommend AISEC university courses 👍 #CyberInternsArmyAustralia #KeepLearning #KeepGoing #KeepGrowingHardik Cholera reacted on thisToday we're launching a new course - Introduction to AI Security. The course is taught by Hemanth Tadepalli, an awesome cyber expert who's brought his AI experience from the autonomous vehicle space to his course. Hemanth has created an outstanding starting point for any AI security learning path. The course has 9 modules, each only 10 minutes, covering these topics: - Data poisoning - Adversarial attacks - Prompt injection - LLM security - Model extraction - Privacy - AI supply chain - Securing AI in production - Governance The course is free and includes certs, CPE credits, and LI badge: https://lnkd.in/gWhS3_ee And please join us Thursday 12pm ET - Hemanth will join me live to discuss the course. https://lnkd.in/ggSknVB6
-
Hardik Cholera shared thisStop 🛑 Think 💭 Act 🎬 #scamsawarenessweek #transurbanHardik Cholera shared thisFor Scam Awareness Week, we are reminding customers to stay vigilant, stop, and slow down when receiving a suspicious text. As they stay alert on the front lines, we are speeding up our response behind the scenes to take the fight directly to scammers. In partnership with Macquarie University, we launched Project RADAR, a new custom-built AI model capable of detecting and taking down malicious SMS scams mimicking trusted brands like Linkt in a matter of minutes. This new technology turns everyday Australians into active defenders. When a customer uploads a screenshot of a suspicious text to us, the AI instantly analyses the message and automates the takedown process with our cybersecurity partners. Since completing the project in May 2026, this system has: • has handled 4,731 reported SMS scams • reported 57 unique fake domains for takedown • boosted our threat reporting response times by 96% • and slashing the average end-to-end domain takedown time to just 4.42 days This success is a testament to the power of joint collaboration. As our Senior Manager of Cyber Defence, Muhammad Ali P., notes "The real-time intelligence gathered enables us to instantly block threats, track emerging scam trends and strengthen our overall Cyber Defence capabilities - effectively disrupting scam operations before they can impact our customers." The core machine learning classifier was engineered at Macquarie University's Cyber Security Hub. Lead researcher Dr Muhammad Ikram highlighted the value of this teamwork "Project RADAR translates high-level academic theory into a tangible, front-line defence mechanism. By automating the extraction, classification, and investigation of threat data, we are dramatically reducing the cost of cyber defence while hitting threat actors where it hurts most - their wallets."
-
Hardik Cholera reposted thisHardik Cholera reposted thisCyber Reserve Force. An interesting "call to arms" for those in Cyber in Australia. I wonder if you get to "hack back"? https://lnkd.in/guJ4ZNdq
-
Hardik Cholera reacted on thisHardik Cholera reacted on thisI just made every OSCP-prep lab from my 050 course free on RatCTF. Here's the reasoning, because it's a deliberate choice, not a fire sale. The biggest thing between someone and OSCP readiness isn't theory — it's reps. Hours in front of real, breakable boxes, building the muscle memory of enumerate → foothold → privesc → repeat. That environment usually costs money, which is exactly the wrong barrier to put in front of someone trying to break into the field. So the labs are open to everyone now. No subscription, no upsell to unlock the "real" content: → Docker-based privesc machines — SUID abuse, cron hijacking, sudo escapes, Linux capabilities → Full attack chains, from enumeration to Active Directory compromise → 96 live machines today, with 8 new themed series (40 machines) landing September 1 → A leaderboard and community hub so you're not grinding alone Honest about what this is and isn't: I can't promise you'll pass. OSCP is OffSec's exam, not mine, and anyone guaranteeing a result is selling you something. What I can give you is the same methodology I teach in the paid course, running on real infrastructure, for free. The labs are free. If you want the structured path wrapped around them — the video walkthroughs, the module-by-module breakdowns, the note templates — that's the paid 050 course: → https://lnkd.in/eZ_j8SiP But you do not need it to start. Register and go: → ratctf.com One rule: these are legal practice grounds. Take what you learn here to authorized targets only. Much love, Wesley (Uncle Rat) #penetrationtesting #bugbounty #cybersecurity #infosec
-
Hardik Cholera reposted thisHardik Cholera reposted this𝗔𝗜 𝗶𝘀𝗻’𝘁 𝘁𝗵𝗲 𝘁𝗮𝗿𝗴𝗲𝘁. 𝗬𝗼𝘂𝗿 𝗱𝗮𝘁𝗮 𝗶𝘀. Six months ago, I shared an infographic mapping the security vulnerabilities emerging around Agentic AI and MCP. At the time, the conversation was largely focused on attacks such as: 𝗣𝗿𝗼𝗺𝗽𝘁 𝗜𝗻𝗷𝗲𝗰𝘁𝗶𝗼𝗻 - manipulating an agent through malicious instructions hidden in content. 𝗖𝗼𝗺𝗺𝗮𝗻𝗱 𝗜𝗻𝗷𝗲𝗰𝘁𝗶𝗼𝗻 - turning an agent’s tool access into direct server-level execution. 𝗧𝗼𝗼𝗹 𝗣𝗼𝗶𝘀𝗼𝗻𝗶𝗻𝗴 - embedding malicious instructions inside trusted tool descriptions. 𝗥𝘂𝗴 𝗣𝘂𝗹𝗹 𝗔𝘁𝘁𝗮𝗰𝗸𝘀 - turning a trusted tool or component malicious through a compromised update. 𝗨𝗻𝗮𝘂𝘁𝗵𝗲𝗻𝘁𝗶𝗰𝗮𝘁𝗲𝗱 𝗔𝗰𝗰𝗲𝘀𝘀 - exposing MCP endpoints without proper authentication. 𝗧𝗼𝗸𝗲𝗻 / 𝗖𝗿𝗲𝗱𝗲𝗻𝘁𝗶𝗮𝗹 𝗧𝗵𝗲𝗳𝘁 - allowing secrets and credentials to leak through logs, files or agent workflows. But the attack surface has expanded. The updated map adds another set of vulnerabilities: 𝗔𝘂𝘁𝗼𝗝𝗮𝗰𝗸 - using an agent’s browser to reach an unauthenticated local MCP endpoint. 𝗔𝗴𝗲𝗻𝘁𝗷𝗮𝗰𝗸𝗶𝗻𝗴 - manipulating coding agents through fake remediation instructions. 𝗭𝗲𝗿𝗼-𝗖𝗹𝗶𝗰𝗸 𝗪𝗼𝗿𝗸𝘀𝗽𝗮𝗰𝗲 𝗧𝗿𝘂𝘀𝘁 - automatically loading MCP configuration when a workspace opens. 𝗧𝗼𝗸𝗲𝗻 𝗣𝗮𝘀𝘀𝘁𝗵𝗿𝗼𝘂𝗴𝗵 - forwarding client tokens to backends without proper validation. 𝗠𝗶𝗱-𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗧𝗼𝗼𝗹 𝗜𝗻𝗷𝗲𝗰𝘁𝗶𝗼𝗻 - hijacking or relabelling tools while an agent is already operating. 𝗠𝗖𝗣 𝗦𝘂𝗽𝗽𝗹𝘆 𝗖𝗵𝗮𝗶𝗻 𝗥𝗖𝗘 - turning trusted components into an attack vector through malicious updates. 𝗧𝗵𝗲 𝗮𝘁𝘁𝗮𝗰𝗸 𝘀𝘂𝗿𝗳𝗮𝗰𝗲 𝗶𝘀 𝗲𝘅𝗽𝗮𝗻𝗱𝗶𝗻𝗴 𝗳𝗮𝘀𝘁𝗲𝗿 𝘁𝗵𝗮𝗻 𝘁𝗵𝗲 𝘀𝗲𝗰𝘂𝗿𝗶𝘁𝘆 𝗺𝗼𝗱𝗲𝗹. From prompts and tools to browsers, workspaces, credentials, sessions and supply chains, every new agentic capability creates another path to exploit. 𝗠𝗼𝗿𝗲 𝗮𝘂𝘁𝗼𝗻𝗼𝗺𝘆 𝗺𝗲𝗮𝗻𝘀 𝗺𝗼𝗿𝗲 𝘁𝗿𝘂𝘀𝘁 𝗯𝗼𝘂𝗻𝗱𝗮𝗿𝗶𝗲𝘀 𝘁𝗼 𝘀𝗲𝗰𝘂𝗿𝗲. And the security question changes from: “𝗖𝗮𝗻 𝘄𝗲 𝘀𝗲𝗰𝘂𝗿𝗲 𝘁𝗵𝗲 𝗔𝗜?” 𝘁𝗼: “𝗖𝗮𝗻 𝘄𝗲 𝗽𝗿𝗼𝘃𝗲 𝗶𝗻𝘁𝗲𝗻𝘁 𝗯𝗲𝗳𝗼𝗿𝗲 𝘄𝗲 𝗴𝗿𝗮𝗻𝘁 𝗮𝗰𝗰𝗲𝘀𝘀?” I’ve updated my original Agentic AI Security Vulnerabilities map with the newer attack patterns. If an AI agent can read, reason, decide and act, should access still be granted 𝗯𝗮𝘀𝗲𝗱 𝗼𝗻 𝗶𝗱𝗲𝗻𝘁𝗶𝘁𝘆 𝗮𝗹𝗼𝗻𝗲? Or should every high-impact action require 𝗽𝗿𝗼𝗼𝗳 𝗼𝗳 𝗶𝗻𝘁𝗲𝗻𝘁? Follow Vinod Bijlani for more insights
-
Hardik Cholera reposted thisWant a career in GRC? Learning these two frameworks is a great place to start. If you learn the NIST CSF 2.0 and ISO/IEC 27001 and understand the controls, you'll be able to help nearly any organization build a stronger security posture. Both of these frameworks have their place in Cybersecurity and this even points out the 20 major differences between the two. You don't always have to be super technical to work in Cybersecurity. GRC is a great place to start. If you don't understand all the controls, that's ok. Lock in one the ones you know and then do a deep dive on the controls you aren't familiar with. If you do that, you'll end up becoming more technical just by deep diving the controls. Save 💾 and put these in your back pocket for later reference. Which framework is normally your go-to?Hardik Cholera reposted thisCybersecurity Excel Dashboard Suite: https://lnkd.in/djn2dTuT Framework or standard? Wrong question. One of the most common cybersecurity governance mistakes is treating NIST CSF and ISO/IEC 27001 as competing choices. They solve different problems. NIST CSF helps organizations structure cybersecurity outcomes, understand current capability, prioritize gaps, communicate risk, and define where improvement should happen. ISO/IEC 27001 brings a formal management-system discipline around that work through defined responsibilities, risk treatment, documented information, governance, assurance, and continual improvement. The real value appears when organizations stop asking: “Which one should we use?” And start asking: “How should we use each one?” A practical approach is to use NIST CSF to create strategic visibility across cybersecurity risk, then use ISO/IEC 27001 to establish the management discipline required to govern, operate, evidence, review, and improve that environment consistently. ✓ One helps organize cybersecurity outcomes ✓ One strengthens management-system accountability ✓ One can expose capability gaps ✓ One can formalize how those gaps are governed ✓ Together they can connect cyber strategy with operational assurance That distinction matters. A cybersecurity program can have excellent controls but weak governance. It can also have strong documentation while failing to prioritize the risks that matter most. Mature security programs need both direction and discipline. They need a way to answer: What outcomes are we trying to achieve? Where are our most important gaps? Which risks deserve priority? Who owns the response? How do we prove the system is operating effectively? How do we keep improving it? That is why framework selection should never become a compliance checkbox exercise. The goal is not to collect standards. The goal is to build a cybersecurity operating model that makes risk visible, decisions defensible, responsibilities clear, and improvement measurable. NIST CSF and ISO/IEC 27001 can be powerful individually. Used intelligently together, they can become even more useful. How does your organization use them: separately, together, or not yet? #CyberSecurity #NISTCSF #ISO27001 #InformationSecurity #GRC #CyberRisk #RiskManagement #ISMS #SecurityGovernance #Compliance #CyberSecurityManagement #InformationSecurityManagement
Hardik Cholera commented on a post
9h
Congratulations Bhavika Unnadkat 👏🏻
-
Hardik Cholera liked thisHardik Cholera liked thisAfter our mobile network outage in July, I committed to sharing what happened, what we'd learnt, and what we'd do differently. Today we're releasing the findings of an external expert investigation. It confirms that the outage was triggered when incorrect date information spread through parts of our mobile network after planned maintenance in our timing system. It also found that we hadn't treated network timing as the critical capability it is, and that ownership and oversight weren't strong enough. We accept the findings, and we've already begun acting on them. Since July we've taken immediate steps to strengthen our timing system, and that work continues. Our commitment is to use what we have learned to make our services even more resilient and reliable for our customers. You can read more on the findings and how we're responding at https://lnkd.in/gvu-x4e7
-
Hardik Cholera liked thisHardik Cholera liked thisAI is rapidly redrawing the cybersecurity workforce—automating entire categories of work and raising the value of human judgement. In my latest Forbes Business Council article, I outline three moves cyber leaders should make to facilitate effective human-machine co-intelligence: 1. Redesign work around human + AI strengths. Automate repetitive work and move people toward judgement, strategy and resilience. 2. Build AI-native security roles. New roles are already emerging around AI supervision, security engineering, governance and assurance. 3. Protect human judgement. As AI becomes more autonomous, leaders must be clear about where machines stop and human accountability begins. Leading CISOs are already redesigning their cyber teams to combine machine intelligence with human judgement, because traditional workforce models no longer hold. My latest Forbes article: Three Moves To Redesign A Cybersecurity Workforce For The AI Era Link in the comments. #Cybersecurity #ArtificialIntelligence #CISO #CyberLeadership
-
Hardik Cholera liked thisHardik Cholera liked thisFolks, As someone who loves disruption and lives for adventure, I’s important to recognize when it’s time for a change… and I couldn’t be more excited about this career inflection point! Let me first start with a few sincerely heartfelt words for some of the most amazing individuals in cybersecurity. Just over 14 months ago, I began what has been an incredible journey with Upwind Security as Field CISO. I joined with unbridled passion for their underdog position in the Cloud Security market after hearing for months about how innovative and CISO-obsessed they were. I leave Upwind today so proud of the exponential growth (in literally every metric possible) that I’ve been a part of, and even more proud of the qualitative ways Upwind is continually applying the practitioner perspective across the board – features, roadmaps, commercials, and more. I’m thrilled to have had a little bit to do with that success, and I’m so excited about where this company is going in the future. I depart Upwind today no less bullish on the importance of bringing runtime/real-time visibility to cybersecurity, of enabling SecOps/DevOps harmony in every organization, and of ensuring simple licensing/billing to CISOs begging to escape the endless SKUs/true-ups. To Rinki Sethi, Amiram Shachar, and the other senior executives: thank you for taking a chance on me and partnering together so strongly... and thanks for a wonderful and rewarding experience. To the rest of the Upwind team: thanks for the friendship and for the awesome run we had together! Until we meet again... and one last time: #UpAndUp! As for what's next... that's for another day. For now, you're just gonna have to wait for me to enlighten you. ;) 🙃 Cheers, Jake #ChiefFunOfficer #WaveMakerOfTheYear
-
Hardik Cholera liked thisHardik Cholera liked thisMajor technology companies including OpenAI, Anthropic, Microsoft, Alphabet and Amazon are urging governments and businesses to strengthen cybersecurity defenses against increasingly capable AI-driven hacking, according to Reuters. In a joint letter issued Thursday, the companies, along with more than 100 others including Broadcom, Capital One, Cloudflare, CrowdStrike, General Motors, IBM, Mastercard, Oracle, Robinhood, Shopify and Visa, said there is limited time to make the digital world more secure before AI-enabled cyberattacks become more widespread. The companies called on governments to expand trusted access programs that give selected organizations access to more powerful AI models ahead of broader public availability. They also urged organizations to make cyber defense an immediate leadership priority. The warning comes as cybercriminals are already using commercial AI tools to assist with attacks. Reuters recently reported that Russian-speaking hackers used SpaceX’s Cursor AI coding assistant to help attack at least seven companies earlier this year. For businesses, the message is straightforward: AI is becoming both a cybersecurity tool and a cybersecurity threat. As AI models become more capable, organizations may need to accelerate investment in AI security, threat detection and defensive automation. What are your thoughts on this? 🤔💬 Source: Reuters (https://lnkd.in/gysQeiEZ) Curious about where AI is headed? 🤖 Follow Future AI to keep your edge over the competition ♻️ Repost if useful
Experience & Education
Licenses & Certifications
Publications
-
Indians and ChatGPT
-
See publicationNewsletter about Indians and ChatGPT.
I've set this newsletter up to bring out best of ChatGPT about 'Indians and ....... (Any relatable topic) & an indian's view on the same topic to bring out 'human' touch to compare and improve Generative AI over time!
Courses
-
Qualys Certified Vulnerability Specialist
-
Languages
-
Hindi
-
-
Gujarati
-
-
English
Full professional proficiency
Organizations
-
AISA
-
-
Cloud Security Alliance
-
Recommendations received
13 people have recommended Hardik
Join now to viewOther similar profiles
-
Sean Macleod
Sean Macleod
Experienced Information Security professional interested in Cloud, DevOps, Big Data, and Machine Learning with a demonstrated history of working in the government, education, health, and banking industries. <br><br>Strong information technology professional skilled in Security Operations, Incident Response, Security Architecture and Design.<br>
Explore more posts
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More