Data Science & AI Engineer
Analyse the data, build the model, ship the AI system.
The hybrid role most companies are actually trying to hire: someone who can interrogate data properly, build a model that holds up, and then take it all the way into a running production system. You go deep on statistics and machine learning, then on deep learning and LLM engineering, and finish by deploying and monitoring what you built.
- Duration
- 20 weeks
- Weekly effort
- 10–12 hrs/week
- Level
- Intermediate
- Format
- Live cohort + project reviews
- Mode
- Classroom in Marathahalli & Live online
- Typical range
- ₹10–35 LPA
Outcomes
What you will be able to do
Not topics covered — capabilities you can demonstrate in an interview and on day one of the job.
- Take a business question, choose the right analysis, and defend your conclusion with evidence
- Build, tune and honestly evaluate machine learning models — and explain them to a non-technical stakeholder
- Train and fine-tune deep learning models, including LLMs, for domain problems
- Engineer AI systems with retrieval, tool use and evaluation, not just notebook prototypes
- Deploy models as services and keep them healthy with monitoring, drift detection and retraining
Curriculum
6 modules over 20 weeks
Every module ends in working code that is reviewed by a mentor before you move on.
01Data Foundations in Python & SQL
Weeks 1–3
Data Foundations in Python & SQL
Weeks 1–3
- Python for data work: clean, testable, modular code rather than notebook sprawl
- pandas and Polars for wrangling messy real-world datasets
- SQL from joins through window functions and query optimisation
- Data cleaning, missing-data strategy and reproducible preparation pipelines
02Statistics & Experimentation
Weeks 4–6
Statistics & Experimentation
Weeks 4–6
- Distributions, exploratory analysis and knowing what the data can and cannot answer
- Hypothesis testing, confidence intervals and the traps that produce false findings
- A/B test design, sample sizing and reading a result honestly
- Introduction to causal inference — separating correlation from cause
- Communicating analysis so it changes a decision
03Machine Learning in Depth
Weeks 7–11
Machine Learning in Depth
Weeks 7–11
- Regression, classification, trees and gradient boosting with XGBoost
- Feature engineering, data leakage, cross-validation and honest evaluation
- Imbalanced data, calibration and choosing the metric that matches the business cost
- Unsupervised learning: clustering, dimensionality reduction, anomaly detection
- Time series forecasting and recommendation systems
- Interpretability with SHAP, plus fairness and bias considerations
04Deep Learning
Weeks 12–14
Deep Learning
Weeks 12–14
- Neural network fundamentals and PyTorch
- Training dynamics: optimisers, regularisation, and debugging a model that will not learn
- CNNs for vision and transformers for text
- Transfer learning and getting real value from pretrained models
05AI & LLM Engineering
Weeks 15–17
AI & LLM Engineering
Weeks 15–17
- Working with LLM APIs: structured output, tool calling and streaming
- Retrieval-augmented generation over your own data with pgvector
- Fine-tuning open-weight models with LoRA, and deciding when not to
- Using LLMs inside a data science workflow: extraction, classification, synthetic data
- Evaluating non-deterministic systems, and guardrails before release
06Production, Monitoring & MLOps Essentials
Weeks 18–20
Production, Monitoring & MLOps Essentials
Weeks 18–20
- Serving models with FastAPI and Docker; batch versus real-time inference
- Experiment tracking, model registry and reproducibility with MLflow
- Cloud deployment and CI/CD for model updates
- Data drift, concept drift and silent failure detection with Evidently
- Retraining strategy, rollback, and the capstone build and defence
Capstone
What you walk out with
Portfolio work substantial enough to carry a technical interview on its own.
An end-to-end AI product: raw data through analysis and modelling to a deployed, monitored service with a live dashboard
A domain problem solved with a fine-tuned or retrieval-augmented LLM, benchmarked honestly against a simpler baseline
Tools & stack
Roles this leads to
- Data Scientist
- AI Engineer
- Machine Learning Engineer
- Applied Scientist
- Decision Scientist
Who it is for
- Analysts and engineers moving into data science and AI roles
- Data scientists who keep getting stuck before production
- Developers who want modelling depth, not just API calls
Prerequisites
- Basic Python and SQL
- School-level mathematics; we rebuild the statistics you need
- No machine learning background required
Enrol
Get the full syllabus for Data Science & AI Engineer
Request the detailed module-by-module syllabus, current batch dates and fee structure. A counsellor will also check whether this is the right programme for your background.
Also consider
Other programmes
Forward Deployed Engineer (FDE)
The engineer who ships AI inside the customer's world.
DetailsGEN-301Generative AI & Agentic AI Engineering
From prompt to production agent — build systems that reason and act.
DetailsFSA-201Full Stack Development with AI
Become the developer who ships complete AI products, alone.
Details