Gen AI Full Stack Engineer (Sprint + Deep Dive)
DIA · Islampur
Become a Gen AI Full Stack Engineer (HighLevel JD aligned) — RAG, agents, cost/latency, eval, AI system design, interview-ready behavioral.
What's included
- 55 video lessons
- 18 hours total
- Level: Intermediate
Instructors
Curriculum
- LLM Foundations for Engineers — The engineering mental model of LLMs: how tokens drive cost, why the context window forces chunking, how temperature shapes determinism,… (6 items)
- Prompting & Function/Tool Calling — The building blocks every downstream system reuses: structured prompting, system vs user roles, structured/JSON output, and function/tool… (5 items)
- RAG Deep Dive — The most important module. Full RAG pipeline: query to embedding to vector search to context retrieval to grounded response. Chunking… (8 items)
- AI Agents & Orchestration — Second most important. Agent architecture: user input to planner to tool calls/APIs to memory to final response. Tool calling in a loop,… (6 items)
- Cost & Latency Engineering — Make AI systems cheap and fast: response/semantic caching, trying smaller models first, model routing, streaming responses, batching,… (5 items)
- Evaluation & Monitoring — You can't ship what you can't measure. Core metrics (hallucination rate, latency, token usage, retrieval accuracy, user satisfaction, task… (5 items)
- AI System Design — Put it all together at scale. End-to-end AI architectures with microservices, queues (Kafka/RabbitMQ/SQS), async processing, retries, rate… (5 items)
- Interview Stories & Product Thinking — Translate engineering into hireable narrative. Ownership framing (take vague requirements, define architecture, ship independently,… (4 items)
- Production Deployment & Scaling — Deep-dive track. Running AI in production: model routing and fallbacks across providers, graceful degradation, blue/green and canary… (5 items)
- Fine-Tuning vs RAG — Deep-dive track. When fine-tuning beats retrieval and vice versa. LoRA and adapters, instruction tuning for behavior/style/task… (5 items)
- Vector Search, Embeddings & RAG (Applied) — How semantic search actually works — embeddings, vector databases, ANN, and HNSW — taught from basics, then applied to the real 'Practice… (7 items)
Enroll in Islampur
Free course — enroll to access all lessons.
Enroll now · Contact DIA for batch schedule.