Experience

3–5 years of industry ML experience

Must-Have Skills

Programming

Strong hands-on programming skills in Python or R (Python preferred)

Ability to write clean, efficient, production-ready code

Applied Machine Learning

Strong problem framing skills:

Choosing between Supervised, Self-Supervised, or Reinforcement Learning

Strong data wrangling & EDA skills, including:

Data preparation and labeling

Feature engineering

Data augmentation techniques

Experience with Weak / Distant Supervision and Pseudo-Labeling

Modeling (End-to-End)

Hands-on experience building models from scratch across:

ML vs DL vs RL

Experience with:

Single models

Ensembles

Mixture of Experts

Strong understanding of optimization techniques (e.g., SGD) and core ML/DL concepts

Transfer Learning

Experience with N-shot / Few-shot learning

Fine-tuning pre-trained models for downstream tasks

ML / DL Specialization

Proven industry or research experience in at least one of:

NLP

Computer Vision

Time Series Modeling

Reinforcement Learning

Frameworks & Libraries

Hands-on experience with TensorFlow or PyTorch

Production & MLOps

Experience deploying ML models into production

Model serving experience using:

TFServing, Seldon, or Custom Serving

Experience with:

Batch, online, and streaming inference

Understanding of performance optimization and scalability

Strong data engineering basics:

Data flow between databases and backend systems

Version Control & Portfolio

Strong Git usage

Active GitHub profile with original ML repositories

Good-to-Have Skills

Experience with MLOps tools:

Kubeflow, MLflow, Airflow, SparkML

Experience designing custom ML algorithms (e.g., custom SGD implementations)

Experience building custom DNN architectures

Kaggle profile with competitive leaderboard rankings

Research publications as first author in reputed ML conferences/journals

AI / Automation patents

Computer Science / IT academic background

Strong foundation in statistics and probability

Experience running Dockerized workloads

Kubernetes exposure (we’re a Kubernetes-first environment)