Stop watching tutorials. Start building real AI systems.
A 3-month, 1:1 mentorship programme for Python developers who want to build production-oriented RAG systems, LangGraph workflows, and MCP integrations through live code review and architecture feedback.
Pankaj Rana — Co-founder & Lead Instructor · View profile →
- 24 sessions, 2×/week
- 3 Months 24 sessions, 2×/week
- All on your GitHub
- 5 Projects All on your GitHub
- No shared cohort
- 1:1 No shared cohort
- Maximum at any time
- 3 Seats Maximum at any time
What Is 1:1 AI Engineering Mentorship?
1:1 AI Engineering Mentorship is a personalised, three-month programme where a single mentor guides one developer through building production RAG systems, LangGraph agents, and MCP integrations — at a pace and curriculum shaped entirely around that developer's goals, existing code, and schedule.
It differs from a scheduled cohort in one key way: there is no fixed group, no fixed syllabus order, and no shared pace. Every session is built around your project and your questions.
You have watched the tutorials. Your GitHub still shows no AI work.
Most developers trying to learn RAG and agents follow the same path: YouTube → Udemy → Medium → tutorial works → try to adapt to a real use case → it breaks → back to YouTube. The loop repeats. The problem is not the content. It is the absence of someone who can see exactly where you are stuck and get you unstuck in real time.
The Tutorial Loop
You replicate the tutorial perfectly. Then you try to adapt it for your own use case and nothing works. Without someone to debug with, most people give up and start the next tutorial.
Architecture Paralysis
LangChain or LlamaIndex? Pinecone or ChromaDB? LangGraph or CrewAI? These decisions matter and affect everything downstream. Without expertise to guide them, developers default to whatever tutorial they last watched.
Empty GitHub
Months of learning, nothing to show for it. No deployable projects. No evidence of what you can build. The portfolio gap is the #1 barrier between mid-level engineers and senior AI engineering roles.
Why 1:1 beats every other learning format for this content
Production AI engineering is not a subject you can learn by following a script. It requires making real decisions, debugging real failures, and building real systems. That requires a real person watching your screen.
Curriculum shaped by your goals
Interview prep? Shipping a specific internal tool? Career pivot? The programme adapts. You are not following a fixed syllabus — you are working toward a specific destination.
Your actual code, debugged live
Group courses debug tutorial code. Here, the mentor looks at your project, your stack, your errors — and explains exactly what went wrong and why, not a generic workaround.
Architecture decisions made together
Should you use Pinecone or Weaviate for your use case? LangGraph or CrewAI? These are decisions engineers need to make independently. You practise making them here, with expert input.
Pace that matches your existing commitments
Working full-time? Busy sprint this week? The schedule flexes. Sessions are reschedulable with 24 hours notice. No cohort to keep up with.
Continuity across all 3 months
Your mentor knows your learning history, your weak spots, and your project context at every session. No repeating yourself, no starting from scratch.
Portfolio that proves what you built
Five production-oriented projects on your GitHub, reviewed against architecture, evaluation, and deployment criteria. Visible technical work, not just a certificate.
What receives individual review
The mentor follows the same project context across all three months. Reviews focus on the decisions and evidence behind your implementation—not on copying a fixed reference solution.
Architecture decisions
Review retrieval, state, tool, data-flow, and deployment choices against the project requirements you can explain and defend.
Code & evaluation evidence
Debug your implementation, define repeatable test cases, inspect failures, and document what the system can and cannot do.
Production-readiness controls
Check permissions, human approval, observability, security tests, incident ownership, and rollback before describing a project as production-oriented.
Curriculum last updated 22 August 2026. Public profile links identify the mentor; employment, credential, client, and outcome claims should be checked against primary evidence before enrolment.
Mentorship vs AI Engineering Course
Both build the same core skills — RAG, agents, and MCP. The right choice depends on whether you want a structured cohort or a fully personalised programme.
| AI Engineering Course | 1:1 Mentorship (this page) | |
|---|---|---|
| Format | Live public cohort | Personalised, just you |
| Schedule | Fixed weekend batch | Flexible — your hours |
| Curriculum | Standard learning path | Custom roadmap |
| Code review | Group project feedback | Every session, your code |
| Pricing | Cohort pricing | Premium fee |
| Best for | Structured peer learning | Personalised acceleration |
Prefer a structured live cohort with peers at a lower price point? See our AI Engineering Course. Hiring on behalf of a team? See Production AI Engineering for teams.
The 3-Month Engineering Journey
Each month ends with deployable projects on your GitHub and measurably deeper skills. Pace adapts to you — milestones are targets, not deadlines.
RAG Mastery — Foundations to Production
- ✓ Build your first RAG pipeline from scratch — PDF chatbot with citations
- ✓ Understand vector embeddings, chunking strategies, and similarity search deeply
- ✓ Set up and compare Pinecone, ChromaDB, and FAISS for your specific use case
- ✓ Implement hybrid search (BM25 + dense) and metadata filtering
- ✓ Add re-ranking to improve retrieval precision measurably
- ✓ Move to Advanced RAG: multi-query, query rewriting, context compression
- ✓ Build a multi-document knowledge assistant with RAGAS evaluation
- ✓ End of Month 1: you have 2 deployable RAG projects on GitHub
Agentic AI — From Pipelines to Autonomous Systems
- ✓ Understand the ReAct framework — the foundation of every modern AI agent
- ✓ Build your first LangGraph agent: nodes, edges, conditional routing, state management
- ✓ Implement memory systems: conversation, entity, and external memory stores
- ✓ Add reflection loops — agents that detect and fix their own errors
- ✓ Multi-agent systems with CrewAI: define roles, tasks, and agent collaboration
- ✓ Compare LangGraph vs CrewAI vs AutoGen — know when to use each
- ✓ Build an AI Research Assistant: supervised multi-step research and report drafting
- ✓ End of Month 2: you have 2 more production-oriented agent projects on GitHub
Tool Calling, MCP & Production Deployment
- ✓ Master function calling across OpenAI, Anthropic, and Gemini APIs
- ✓ Build agents with real integrations: Gmail, Slack, Calendar, and database querying
- ✓ Understand MCP (Model Context Protocol) — architecture and enterprise use cases
- ✓ Build a custom MCP server from scratch, exposing tools to Claude and Cursor
- ✓ Design least-privilege MCP and tool permissions with human approval for write or high-impact actions
- ✓ Test prompt injection, sensitive-data disclosure, output handling, and tool-abuse failure modes
- ✓ Add observability, audit events, prompt and model versioning, and production evaluation gates
- ✓ Deploy your capstone to a cloud endpoint (AWS Lambda, GCP Cloud Run, or Vercel)
- ✓ Create a system card, risk register, evaluation report, incident runbook, and rollback plan
- ✓ End of Month 3: capstone demonstration + production-readiness and portfolio review
Production Readiness & Responsible AI Review
Building a working demo is only the first milestone. During code and architecture reviews, each project is assessed against governance, security, evaluation, human-oversight, and operational acceptance criteria so you can explain both how the system works and how its risks would be managed.
Purpose, Data & Ownership
Define intended use, users, limits, data flows, access boundaries, system owner, approver, and escalation path before treating a prototype as production-oriented.
Evaluation & LLM Security
Build repeatable tests for retrieval, groundedness, failure cases, prompt injection, sensitive-data disclosure, improper output handling, and unsafe tool use.
Permissions & Human Oversight
Apply least privilege to tools and MCP servers, validate outputs, and place human approval before write, external-communication, or other consequential actions.
Operations & Recovery
Version models and prompts, record useful audit events, monitor quality and failures, assign incident ownership, and document rollback and decommissioning steps.
Capstone evidence pack
What your readiness review produces
These engineering artefacts make your assumptions, controls, test evidence, limitations, and response plan visible to a technical reviewer rather than leaving them implicit in the code.
- System card: intended use, limitations, owners, and human checkpoints
- Threat model and concise risk register with control owners
- Tool and data access matrix based on least privilege
- Evaluation report with thresholds, failures, and residual risks
- Launch checklist and runbook for monitoring, incidents, and rollback
Reviews are informed by lifecycle practices in the NIST AI Risk Management Framework and application risks in the OWASP Top 10 for LLM Applications. Continue with our guides to AI governance, responsible AI, and LLM evaluation. Mentorship does not provide certification or legal or regulatory approval.
5 Projects. All on Your GitHub.
Every project is designed around production-oriented engineering criteria. Each becomes a portfolio piece you can explain in technical interviews and adapt further for an approved company environment.
PDF Chatbot
Month 1, Week 2LangChain · ChromaDB · OpenAI Embeddings
Company Knowledge Assistant
Month 1, Week 4LlamaIndex · Pinecone · RAGAS Evaluation
AI Research Assistant
Month 2, Week 6LangGraph · ReAct · Tool Calling
Multi-Agent Workflow System
Month 2, Week 8CrewAI · AutoGen · Memory Persistence
Custom MCP Server + Deployed Capstone
Month 3, Week 12MCP · Claude Integration · Cloud Deployment
All 5 projects committed to your personal GitHub. All code is yours — no licences, no restrictions.
Who This Is For (and Who It Isn't)
- ✓ Python developer with 2–8 years of experience
- ✓ Has used ChatGPT or Claude — understands what LLMs can do
- ✓ Wants to build AI systems, not just use AI tools
- ✓ Has a specific goal: job switch, internal project, or startup idea
- ✓ Can commit 8–10 hours per week including session time
- ✗ Complete Python beginner (no coding background)
- ✗ Looking for a quick certificate without building real projects
- ✗ Wants purely business/non-technical AI training
How the Process Works
Application & Fit Call
Fill out the enquiry form. We schedule a 30-minute call to understand your background, goals, and whether this programme is the right fit. No hard sell — if it is not right for you, we will say so.
Technical Assessment
A short async task to gauge your current Python level and how you approach problems. Takes 45–60 minutes. Not a pass/fail — it helps us design your Month 1 starting point.
Personalised Roadmap
Before Session 1, you receive a customised 3-month roadmap aligned to your specific goals — whether that is a job switch, shipping an internal tool, or building a portfolio.
Session 1 — First Build
No separate orientation session. Session 1 starts coding immediately, with a first working RAG pipeline as the initial target. Progress depends on your starting code and environment setup.
Ongoing Sessions + Async Support
Two sessions per week, with async code review, resource sharing, and questions answered between sessions via WhatsApp or Slack.
Month-End Reviews
At the end of each month, a dedicated review session assessing your GitHub projects, identifying gaps, and adjusting Month 2/3 focus accordingly.
What Past Mentees Built
“ I had been watching YouTube tutorials on LangChain for six months and building nothing I could actually show. Three months with a mentor who reviewed my actual code changed that completely. I now have 4 projects on GitHub that are genuinely production-quality. ”
Rahul Sharma
Senior Software Engineer → AI Engineer
Moved to a product company AI role
“ The MCP module in Month 3 was the career differentiator I wasn't expecting. I was the only person in my company who knew how to build an MCP server. That visibility led directly to a promotion to the AI architecture team. ”
Divya Krishnan
ML Engineer
Bangalore-based AI startup
“ Best learning investment I've made. The 1:1 format meant every session was exactly what I needed — no time wasted on things I already knew, and deep dives when I was stuck. My RAG evaluation scores went from 58% to 91% in 6 weeks. ”
Karthik Menon
Data Scientist → AI Systems Engineer
Chennai-based fintech
Investment & What You Get
- ✓ 24 live 1:1 sessions (2 hrs each) — 48 hours of dedicated instruction
- ✓ Pre-programme technical assessment + personalised roadmap
- ✓ Async code review and Q&A between sessions (WhatsApp/Slack)
- ✓ All 5 portfolio projects — your code, your GitHub, yours to deploy
- ✓ Month-end reviews with adjusted focus for the following month
- ✓ LangGraph agent architecture reference guide
- ✓ Custom MCP server starter kit
- ✓ RAG evaluation dashboard template (RAGAS integration)
- ✓ Technovids AI Engineering certificate on completion
- ✓ 30-day post-programme support after Month 3 ends
💡 Your company should pay for this. The projects you build are directly applicable to your employer's AI initiatives. Many mentees expense this as professional development. We can provide an invoice addressed to your company.
3-Month 1:1 Mentorship
₹75,000
All-inclusive · 24 sessions · 5 projects
(₹25,000 per month or single payment)
Availability
Maximum 3 mentees at any time — by design. If slots are full, you will be added to a waitlist and contacted when a slot opens (typically 4–6 weeks).
Mentorship refund terms are confirmed in writing before enrolment. Review the Refund Policy.
Frequently Asked Questions
How is 1:1 mentorship different from the group corporate course? +
The group course delivers a fixed curriculum to a team over 5 days or 8 weeks. The 1:1 mentorship is paced around your learning speed, project ideas, career goals, and existing code. Sessions can focus on the architecture or implementation problem you are currently working through.
What is the time commitment per week? +
Two sessions per week, each 2 hours, plus approximately 4–6 hours of project work between sessions. The total expected commitment is roughly 8–10 hours per week and is designed to fit around full-time work.
What Python level do I need? +
You should be comfortable writing Python functions, working with classes and dictionaries, and making API calls. Prior experience with async Python or an LLM framework is not required. If you have built a simple REST API or used pandas or requests, you have the expected foundation.
Can I use this to prepare for senior AI engineering roles? +
The programme can support preparation for senior AI engineering roles by helping you build and explain five production-oriented GitHub projects. RAG, stateful agent workflows, evaluation, deployment, and MCP integrations are relevant to many AI engineering roles, but job requirements and hiring outcomes vary by company.
How many mentees do you take at one time? +
The programme is limited to a maximum of three active mentees so that each participant receives dedicated code review and architecture feedback. If all places are occupied, new applicants may be offered a waitlist place.
What happens if I need more time on a topic? +
The pace is flexible. The curriculum provides a roadmap, but sessions can spend additional time on topics such as LangGraph state management, retrieval evaluation, or deployment when your project requires it.
Is the programme available online only, or can it be in-person in Bangalore? +
Sessions are primarily delivered online through Zoom or Google Meet. In-person sessions in Bangalore may be available on request and are confirmed during the clarity call.
What is the refund policy? +
Refund and cancellation terms for the 1:1 mentorship are confirmed in writing before enrolment. Review the Refund Policy and ask about mentorship-specific terms during the clarity call before making a payment.
Is AI governance and LLM security included in the mentorship? +
Yes. Governance and security are applied to the portfolio projects through system ownership and risk classification, data and access boundaries, prompt-injection and sensitive-data testing, least-privilege tool permissions, human approval, evaluation, audit logging, monitoring, incident response, and rollback planning.
What makes the portfolio projects production-oriented? +
Each project is reviewed against architecture, evaluation, security, permissions, human-oversight, observability, and recovery criteria. You create a system card, risk summary, evaluation evidence, and an operating runbook for the capstone. Production-oriented does not mean certified, compliant, or approved for uncontrolled use; organisation-specific security, privacy, and legal review is still required.
AI Engineering Knowledge Base
New to AI engineering? These guides cover the core concepts you will build into production systems during the mentorship.
Apply for the Mentorship Programme
Tell us about your background and what you want to build. We will reach out within 24 hours to schedule a 30-minute fit call.