Generative AI, Large Language Models (LLM) & Agentic AI Full Stack Engineering

Hardware, OS, processes, threads, memory, networking, client-server models, web architectures, and developer tooling.

Python syntax, dynamic typing, primitive and collection types, operators, conditionals, loops, functions, and string formatting.

Object-oriented design, inheritance, magic methods, exception handling, custom exceptions, modules, and file handling.

Comprehensions, functional programming, iterators, generators, decorators, type hints, dataclasses, and async programming.

Version control fundamentals, branching, merging, rebasing, pull requests, code reviews, and GitHub Actions.

Time/space complexity, arrays, strings, linked lists, stacks, queues, trees, heaps, graphs, recursion, and interview problems.

RDBMS principles, CRUD operations, joins, subqueries, CTEs, window functions, database design, indexing, and transactions.

PostgreSQL installation, schemas, JSONB, full-text search, CTEs, triggers, and AI conversation/metadata storage.

HTML5 semantic markup, forms, modern CSS, Flexbox, Grid, responsive layouts, and AI application UI interfaces.

JavaScript fundamentals, ES6+ features, DOM manipulation, asynchronous promises, Fetch API, and streaming responses.

TypeScript types, interfaces, generics, utility types, and strict domain modeling for AI applications.

React 19 component architecture, hooks, routing, global state, API integration, and production AI application UIs.

FastAPI architecture, ASGI, Pydantic validation, dependency injection, middleware, streaming responses, and WebSockets.

Django MVT, ORM QuerySets, migrations, admin portal, serializers, ViewSets, and enterprise admin systems.

REST principles, HTTP status codes, headers, pagination, filtering, versioning, rate limiting, and AI-specific API design.

User registration, password hashing, JWT access/refresh token rotation, role-based access control (RBAC), and OAuth2.

AI taxonomy, machine learning paradigms, end-to-end ML workflows, classical algorithms, and evaluation metrics.

Linear algebra, matrix operations, probability, Bayes theorem, statistics, calculus, and gradient descent for AI systems.

Multidimensional arrays, vectorization, Pandas Series/DataFrames, data cleaning, aggregation, and AI data preprocessing.

Preprocessing, supervised models, hyperparameter tuning, model evaluation, and practical candidate skill classification.

Neural networks, perceptrons, activation functions, loss functions, backpropagation, and PyTorch with CUDA acceleration.

Text preprocessing, tokenization, stemming/lemmatization, POS tagging, NER, TF-IDF, word embeddings, and text classification.

Transformer architecture, encoder-decoder, self-attention, multi-head attention, positional encodings, and model families.

LLM pretraining, tokenization, context windows, reasoning, capabilities, failure modes, hallucinations, and open-weight models.

API keys, authentication, inference parameters, streaming, retries, rate limiting, cost tracking, and reusable LLM service design.

Instructions, context, zero/few-shot prompting, Chain-of-Thought, structured JSON outputs, Pydantic validation, and security.

Dense vector embeddings, semantic similarity, distance metrics, chunking strategies, and end-to-end vector pipelines.

Vector indexing, pgvector, Chroma, Qdrant, FAISS, vector CRUD, metadata filtering, hybrid search, and HNSW graph indexing.

RAG lifecycle, document parsing, fixed vs semantic chunking, vector retrieval, context injection, and citation attribution.

Query rewriting, multi-query expansion, HyDE, cross-encoder reranking, context compression, and Ragas evaluation.

PDF ingestion, OCR for scanned documents, table detection, form extraction, and multi-page document processing pipelines.

Autonomous agents vs chatbots, planning, reasoning, tool use, agent state, execution loops, and self-correction.

Tool definitions, JSON schemas, function execution, argument validation, custom tool building, and sandboxing.

Agent workflows, ReAct pattern, router pattern, supervisor pattern, and multi-agent team orchestration.

MCP architecture, Client-Server model, tools, resources, prompts, custom server implementation, and security.

Short-term, conversation, long-term, semantic, and episodic memory with multi-tier storage in Redis, Postgres, and vector stores.

Text, image, audio, video, vision understanding, OCR, Speech-to-Text (STT), Text-to-Speech (TTS), and multimodal interview systems.

Code generation, explanation, refactoring, bug detection, AST chunking, symbol-aware indexing, and repository RAG.

Benchmark methodologies, human evaluation, LLM-as-a-judge, Ragas metrics for RAG, and agent evaluation frameworks.

Token usage telemetry, inference latency, cost tracking, distributed tracing, span trees, and production monitoring.

OWASP Top 10 for LLMs, prompt injection, jailbreaks, data leakage, PII masking, tool execution guardrails, and sandboxing.

AI safety, privacy, bias mitigation, fairness, transparency, explainability, human-in-the-loop oversight, and governance.

Token economics, prompt optimization, semantic caching with Redis, intelligent model routing, and batch processing.

Production architecture design, React, API Gateway, FastAPI, vector stores, Redis, PostgreSQL, Kafka, and fault tolerance.

Conversation buffers, session management, rate limiting, semantic caching, agent state, and distributed locks.

Asynchronous processing, document extraction, embedding generation, task queues, workers, and Celery Beat.

Service boundaries (Auth, Document, RAG, Agent, Eval), inter-service REST/gRPC/Kafka communication, and resilience.

Event-driven AI pipelines, document processing events, evaluation events, Kafka producers, consumer groups, and scaling.

Enterprise Linux administration, filesystem, permissions, processes, networking, SSH bastions, and Bash automation.

Containers, Docker CLI, multi-stage Dockerfiles for Python/AI, Docker Compose orchestration, and security hardening.

Kubernetes cluster architecture, control plane, pods, deployments, services, ingress, autoscaling, and operations.

DevOps lifecycle, continuous integration, continuous delivery, automation, infrastructure as code, and monitoring.

Workflows, jobs, steps, runners, automated testing, security scanning, container builds, and deployment to Amazon ECR/EKS.

Jenkins architecture, controller, agents, declarative pipelines, Jenkinsfile, plugins, credentials, and webhooks.

Infrastructure as Code, HCL, providers, resources, variables, modules, state management, and provisioning AWS infrastructure.

Cloud computing models, AWS global infrastructure (Regions, AZs), IAM users, groups, roles, policies, and least privilege.

EC2 instances, AMIs, EBS storage, security groups, Elastic IPs, application deployment, Nginx reverse proxy, and SSL.

VPC design, CIDR blocks, public/private subnets, route tables, IGW, NAT Gateways, security groups, NACLs, and endpoints.

S3 buckets, objects, storage classes, lifecycle policies, encryption, presigned URLs, and storing training documents/resumes.

Managed PostgreSQL/MySQL, automated backups, multi-AZ, read replicas, and storing users, conversations, and AI metadata.

NoSQL tables, partition/sort keys, query/scan operations, GSI, TTL, streams, and storing agent state and activity logs.

Serverless architecture, Python runtimes, handlers, layers, triggers, cold starts, and serverless document processing.

REST/HTTP APIs, routes, integrations, authentication with Cognito/JWT, throttling, CORS, and deployment stages.

Standard and FIFO queues, visibility timeouts, dead-letter queues, retries, and asynchronous AI document processing.

SNS topics, fan-out architectures, EventBridge rules, event buses, and asynchronous AI event-driven workflows.

Elastic Container Registry, EKS cluster setup, node groups, IAM roles for service accounts, and scaling AI microservices.

Managed foundation models (Claude, Llama, Titan), Bedrock Knowledge Bases, Bedrock Agents, and Python Boto3 integration.

Metrics, log groups, alarms, operational dashboards, CloudTrail audit trails, and monitoring token usage and latency.

MLOps lifecycle, data pipelines, model registries, experiment tracking, model deployment, and continuous monitoring.

Prompt, model, and dataset versioning, golden test datasets, prompt regression testing, and quality/cost/latency monitoring.

End-to-end AI CI/CD pipeline, automated prompt/RAG/agent test suites, safety tests, and automated model promotions.

Security in CI/CD, SAST, DAST, secret scanning, container scanning, AI threat modeling, and prompt injection testing.

System design for complete AI platforms: users, frontend, API gateway, FastAPI, RAG, agents, vector DBs, and scalability.

Async processing, connection pooling, RAG chunk/rerank optimization, token reduction, streaming, and autoscaling.

Deep-dive implementation review of 8 industry production projects from AI Chat and RAG to Autonomous Agents and MCP platforms.

MockAttempt AI Career Intelligence Platform: candidate profiling, skill analysis, RAG curriculum assistant, and AI interviewer.

Unit/integration testing, API testing, prompt behavior testing, structured output testing, and golden dataset regression.

Evaluation metrics for student outputs, agent multi-step execution, tool selection, RAG faithfulness, and user feedback.

Responsible AI, privacy, data protection, PII handling, access control, audit logs, model governance, and compliance.

Comprehensive technical interview prep across Python, FastAPI/Django, React, GenAI, Transformers, RAG, Agents, AWS, and DevOps.

ATS optimization, quantified achievement statements, GitHub repository standards, live AI demos, and portfolio website.

5-Level multi-modal assessment architecture: Knowledge MCQs, Technical Analysis, Practical Tasks, Project Reviews, and AI Vivas.

Distinguishing course completion from demonstrated skill evidence, 7 competency dimensions, and verified skill passport.

16-stage progressive verification pipeline, mandatory gate clearances, and placement assurance activation.

Course hierarchy (Course -> Module -> Topic -> Lesson), multi-modal learning assets, and integrated verification gates.