Dr. Dipen
I am an AI/ML researcher with 150+ citations and 16 published research papers. I have three tier-1 publications, including Internet of Things (Elsevier), Biomedical Signal Processing and Control (Elsevier), and IEEE Access. In my research journey, I have collaborated with NASA Glenn Research Center, Cleveland Clinic, and the U.S. Department of Energy for various research projects. I am also an official reviewer and have reviewed over 100 research papers for Elsevier, IEEE Transactions, ICRA, MDPI, and other top journals and conferences. I hold a PhD from Cleveland State University with a focus on large language models (LLMs) in cybersecurity, and I also earned a master’s degree in informatics from Northeastern University.
Contributed IDE list
No IDE contribution for now
User's comments
articles
Install GraphQL for n8n Workflows That Call an LLM During Agent Orchestration
Your agent only knows what you put in its prompt. Choosing which API fields land there stops being optional the moment you connect an LLM to real data, because every extra field is text the model reads and you pay for. GraphQL gives you that choice inside an n8n workflow. With field selection, your…Nov 20th 2026
Build an AI Agent API in Rust: An Actix Tutorial for Tool Calling, RAG, and Fast LLM Backends
Rust's Tool trait catches JSON schema mismatches at compile time—before you ship. Actix's async runtime handles the overhead when you split agents across multiple LLM calls and stream tokens back. Rig's cassette-based testing replays recorded provider responses during load tests, so you validate…Nov 19th 2026
madrl vs MRKL: Which Agent Design Fits Real AI Applications Better
Your agent architecture stopped being a research question the moment you put it behind a product with real users. MRKL and MADRL put the work in different places, and you pay for that placement on every release. A MRKL system puts the effort into engineering modules you can read. MADRL (multi-agent…Nov 19th 2026
AWQ Quantization vs 8-Bit and 4-Bit: What Actually Cuts LLM Inference Cost Without Breaking Quality
Inference cost stops being a notebook problem the moment you serve a model on hardware someone pays for every month. In FP16 (16-bit floating point, the default weight format), a 32B model needs 60+ GB of GPU memory. A 70B model needs about 140 GB for weights alone—more than a single 80 GB GPU…Nov 18th 2026
AI Log Best Practices for RAG Apps: What to Track, Store, and Redact
AI logging stopped being optional the moment you moved your RAG prototype into production. Most enterprise LLM systems now run some form of RAG (retrieval-augmented generation, the pattern where you fetch relevant documents and feed them to the model alongside the prompt). Most ship without the…Nov 18th 2026
12 Prompt Engineering Examples That Hold Up in Real LLM Apps, From Few-Shot Prompts to RAG
More than 60% of organizations are building AI-powered retrieval tools specifically to reduce hallucination rates and personalize outputs with internal data. That number matters because RAG doesn't eliminate hallucination — it shifts where the risk lives. Models can still generate false statements…Nov 17th 2026
LLM Selection Drives Agent Reliability More Than Prompt Engineering
Watch: Lesson 7 | Prompt Engineering, Model Selection & Validator Agents by RhemaAI Systems Model selection sets the ceiling on agent reliability. Prompt engineering and guardrails help you reach that ceiling, but they can't raise it. When a model lacks the reasoning capacity for your task, no…Nov 17th 2026
Reactive vs Proactive vs Hybrid Agents: Pick by Task Complexity
The short version: A fine-tuned Qwen2-7B-Instruct model performed better than every baseline on proactive assistance tasks. Reactive agents work when you need millisecond response times and zero state, but fall apart the moment context spans multiple turns. Proactive behavior is a learned skill,…Nov 16th 2026
How to Build AI Agents: Write Evals First, Code Second
Watch: LLM as a Judge: Scaling AI Evaluation Strategies by IBM Technology Write 20 to 50 evals from real failures before you scale complexity. Build them as a five-column grid: workflow step, starter test-set size, what you measure, success metric, iteration loop. Two rules anchor the approach.…Nov 16th 2026
Promptfoo vs DeepEval for AI Search: Which Finds Bad Prompt Patterns Faster in RAG Apps
In short: Most production teams run both tools because they catch different failures. Promptfoo shows evaluations as a matrix where rows are test cases and columns are model configurations—each cell shows pass/fail. DeepEval gives you research-backed RAG metrics: ContextualRecallMetric,…Nov 15th 2026
Mastering LLM Fine-Tuning and Inference Techniques
The short version: Building AI agents that work in production means mapping your fine-tuning approach to the data and infrastructure you actually have. Teams stall out when they pick a technique because it sounds advanced rather than because it fits their constraints. Parameter-efficient…Nov 15th 2026
Reactive, Deliberative, and Multi-Agent: Choosing the Right Agent Type Before Writing Code
The choice between Reactive, Deliberative, and Multi-Agent patterns isn't a progression. These are different decision models. Pick the wrong one, and you pay in latency, tokens, and debugging time. Reactive agents respond to current input without memory or planning. Think security alerts that…Nov 14th 2026
dotenv in Node for AI Coding Tools: Safe API Key Setup for OpenAI, Anthropic, and MCP Apps
Every AI key you paste into a source file is one git push away from going public. Setting up dotenv in Node takes a few minutes. dotenv is a small package that moves those keys into a local .env file, and your code reads them at runtime through process.env (Node's built-in object for environment…Nov 14th 2026
Build Your First AI Agent: Define the Tool Contract Before the Loop
{"article":" \n\n## Why the tool contract comes before the loop\n\nThe tool contract stopped being optional the moment you tried to ship an assistant that does real work. Skip it and you get loops that never terminate, tools called with malformed arguments, and a token bill that grows every turn.…Nov 13th 2026
Examples of Reinforcement Learning in LLM Apps: From RLHF to Safer Business Outputs
Watch: Reinforcement Learning from Human Feedback (RLHF) Explained by IBM Technology In our experience, preference-tuned models can outperform much larger unaligned counterparts on human preference evaluations — alignment work often matters as much as raw scale when the target is human preference.…Nov 13th 2026
Reinforcement Learning in Python for LLM Fine-Tuning: Build a Simple RLHF PPO Pipeline
Short version: RLHF with PPO starts from a supervised-fine-tuned model and nudges it toward preferred outputs. You train a reward model on pairwise comparisons, then use PPO to update the policy while keeping it tethered to the starting checkpoint. The training loop generates completions, scores…Nov 10th 2026
Python dotenv for LangChain Agents: Safe .env Patterns for RAG, Tools, and Local LLM Apps
Watch: Build a RAG (Retrieval-Augmented Generation) System from Scratch | Python + Vector DB + LLM by CodeVisium Python-dotenv gives local LangChain projects a simple way to keep credentials out of source code while still making them available through os.environ. It works well for solo prototypes,…Nov 9th 2026
Let Your Internal AI Agent Rewrite Its Own Prompts From Failed Runs
When you build AI agents that rewrite their own prompts after failed runs, you're addressing a gap that's been draining engineering time: the hidden orchestration tax of manual prompt iteration. The shift matters because every ambiguity in an agent prompt compounds across steps, turning vague…Nov 6th 2026
Grep Beats Embeddings for Some AI Coding Tool Tasks: Use Deterministic Search Before You Build RAG
Watch: Is RAG Still Needed? Choosing the Best Approach for LLMs by IBM Technology Short version: A SWE-Bench Verified agent using only grep and find reached the top of the leaderboard without any semantic indexing. A single exact-name query can use far fewer tokens with grep than broad embedding…Nov 3rd 2026
Build an AI Agent in 60 Lines of Python Before Reaching for LangChain
Watch: Build an AI Agent From Scratch in Python - Tutorial for Beginners by Tech With Tim Understanding the agent loop stopped being optional the moment you tried to change one thing in a framework demo and couldn't find where the loop lived. Quickstarts run fine until you need to touch them. A…Nov 2nd 2026
courses
AI Accelerator
Land an AI engineering role in as little as 90 days, without going back to school, grinding through YouTube tutorials, or needing any prior AI experience. We build your personalized roadmap, help you build a production-grade portfolio, apply for jobs on your behalf, and prep you for interviews, all the way through to a signed offer. If you don't land a role within 6 months of us starting to apply on your behalf, you get 100% of your tuition back.Jul 11th 2025
AI bootcamp 2
This advanced AI Bootcamp teaches you to design, debug, and optimize full-stack AI systems that adapt over time. You will master byte-level models, advanced decoding, and RAG architectures that integrate text, images, tables, and structured data. You will learn multi-vector indexing, late interaction, and reinforcement learning techniques like DPO, PPO, and verifier-guided feedback. Through 50+ hands-on labs using Hugging Face, DSPy, LangChain, and OpenPipe, you will graduate able to architect, deploy, and evolve enterprise-grade AI pipelines with precision and scalability.Aug 12th 2025
books
Dipen hasn't published any books