Mehrab Robin

Directions Website

Mehrab Robin Engineer | Technologist | Researcher I managed digital broadcasting and social media strategies to ensure secure and authentic content delivery.

Tech Contributor | IT Consultant | Digital Presence Expert | WebTech | Blockchain | Digital Forensic | Cybersecurity | Signal Processing | AI & ML

Since 2014, I have been working as a professional in web development, digital growth, and enhancing online presence. As an Operations and Systems Engineer at Channel i, one of Bangladesh’s leading television networks, I focused on web security, digital rights management, and content piracy protection. Additionally, my freelance journey allowed me to support over 50 organizations worldwide in optimizing their digital presence and growth. I’m now pursuing a Ph.D. in Cyberspace Engineering at Louisiana Tech University, specializing in digital forensics and blockchain security. My research interests include digital transformation, data science, machine learning, artificial intelligence, and computer vision. Previously, I was awarded a fully-funded scholarship from Delhi Technological University, where I completed my M.Tech in Signal Processing and Digital Design with a focus on Machine Learning. My diverse academic and professional experiences inspire my passion for advancing digital security and technology.

Writing With AI Without Losing Your VoiceGenerated prose is fluent and anonymous because models produce the typical cont...
09/18/2026

Writing With AI Without Losing Your Voice

Generated prose is fluent and anonymous because models produce the typical continuation and voice is the atypical one. What preserves a voice in practice....

Why AI-assisted prose sounds anonymous, the constructions that give it away, and how to use AI as an editor rather than an author.

Post-Quantum Cryptography on Embedded and IoT DevicesA sensor shipped in 2026 will outlive RSA. Why signature sizes brea...
09/14/2026

Post-Quantum Cryptography on Embedded and IoT Devices

A sensor shipped in 2026 will outlive RSA. Why signature sizes break radio budgets, which primitives fit, and what to demand from vendors now....

Why post-quantum signatures are hard on constrained devices, which primitives fit an embedded budget, and the questions to put to vendors.

Evaluating AI Agents: Accuracy Is the Wrong MetricAgents have no single correct path, succeed partially, and behave diff...
09/08/2026

Evaluating AI Agents: Accuracy Is the Wrong Metric

Agents have no single correct path, succeed partially, and behave differently run to run. What to measure instead of accuracy, and why failure shape matters most....

Why accuracy fails as an agent metric, and what to measure instead: outcome completion, cost per success, failure modes and variance across runs.

Why Two GPUs Are Not Twice One GPUMulti-GPU scaling disappoints because the interconnect, not the silicon, is usually th...
09/04/2026

Why Two GPUs Are Not Twice One GPU

Multi-GPU scaling disappoints because the interconnect, not the silicon, is usually the limit. When more cards help, and when one bigger card is the answer....

Why adding a second GPU rarely doubles performance, how data and model parallelism differ, and when one larger card is the better purchase.

Where AI Actually Saves Time in Research, and Where It Costs YouA working rule from a PhD student: use AI where being wr...
08/31/2026

Where AI Actually Saves Time in Research, and Where It Costs You

A working rule from a PhD student: use AI where being wrong is cheap and detectable, avoid it where being wrong is expensive or invisible....

Where AI genuinely speeds up research work, where it quietly costs time, and a simple rule for deciding which situation you are in.

A Security Model for AI Agents That Can Use ToolsYou cannot prevent an agent from making a bad decision. You can make su...
08/25/2026

A Security Model for AI Agents That Can Use Tools

You cannot prevent an agent from making a bad decision. You can make sure a bad decision cannot do much. A practical security model built around tools, not prompts....

How to secure AI agents with tool access: least-privilege tools, separating reading from acting, human confirmation and structured audit logs.

Building a Cryptographic Inventory: The Step Everyone SkipsYou cannot migrate cryptography you cannot find. How to build...
08/21/2026

Building a Cryptographic Inventory: The Step Everyone Skips

You cannot migrate cryptography you cannot find. How to build an inventory that covers the appliances, firmware and vendor products where most of it actually lives....

A practical method for finding every place RSA and ECC live in your estate, and how to prioritise the migration by data lifetime.

Hybrid Key Exchange in TLS: The Post-Quantum Step You Can Take TodayCombining an elliptic curve exchange with a post-qua...
08/17/2026

Hybrid Key Exchange in TLS: The Post-Quantum Step You Can Take Today

Combining an elliptic curve exchange with a post-quantum one is secure unless both break. It is widely supported already, and often just a setting at your CDN....

How hybrid post-quantum key exchange works in TLS, why it beats waiting, and what to check before rolling it out across your estate.

Not a Ladder: What OpenAI’s Own Numbers Say About Sol, Terra and LunaOpenAI's published tables put Terra and Luna within...
08/13/2026

Not a Ladder: What OpenAI’s Own Numbers Say About Sol, Terra and Luna

OpenAI's published tables put Terra and Luna within a tenth of a point on professional agentic work and 48 points apart on long-context recall. Route by failure mode, not difficulty....

OpenAI's own tables show Terra and Luna nearly tied on agentic work and 48 points apart on long context. Route GPT-5.6 by failure mode, not difficulty.

Prompt Injection Is the New SQL Injection — Without the FixSQL injection was solved by separating code from data. Langua...
08/11/2026

Prompt Injection Is the New SQL Injection — Without the Fix

SQL injection was solved by separating code from data. Language models cannot make that separation, which is why prompt injection defences have to be architectural....

Why prompt injection resists the defences that fixed SQL injection, and which architectural controls actually bound the damage when it succeeds.

Address

Batchelor, LA

Website

, ,

Alerts

Be the first to know and let us send you an email when Mehrab Robin posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Shortcuts

Share

Category