Saba Iram N
Python Developer
1+ project
Python
JavaScript
SQL
Node JS
Active 1 day ago
Invite to interview
Download CVCV
Overview
Technical skills
Projects
Timeline
Roles
Overview
Generalist developer at a junior level focusing on web application development and basic NLP/transformers demonstrations. The strongest proven skill is building Django CRUD web applications as shown by implemented settings, views and forms code. Not evidenced are custom model training, experiment tracking, production ML serving, or performance optimization.
Phone
Technical skills
Languages
4
Python
JavaScript
SQL
Node JS
Databases
3
MongoDB
PostgreSQL
MySQL
AI/ML
5
NumPy
OpenCV
huggingface_hub
CUDA
Hugging Face
Frontend
3
React.js
Tailwind CSS
Bootstrap
DevOps
3
GitHub
Netlify
Rest API
Other
9
Requests
Express
Django
Machine Learning
AI Agents
PostCSS
NLP
Post-training
Function Calling
Projects
Django CRM Application
Created a CRM application using Django for the backend and MySQL for data storage. Used Bootstrap for the UI styling and Django components to support core CRM functionality. Focused on building a structured web app with model-driven data management.
Developed an admin dashboard UI using React and JavaScript with Tailwind CSS styling. Implemented administrative screens and interactive front-end components for managing application content. Focused on clean dashboard layout and responsive usability.
Built a notes web application using a React frontend with a Node.js and Express.js backend. Used Tailwind CSS for UI styling and implemented full CRUD-style note management backed by a database layer. Included a feedback component for user input.
Timeline
National Institute of Engineering Mysore (NIE Mysore)
Bachelor's Degree
Python Developer Intern
•
Junior
iRobo Automation Pvt. Ltd.
•
Internship
Worked as a Python intern supporting analysis based on chemical and physiological features. Assisted with handling datasets that include cases, recoveries, and deaths. Contributed to feature-related work and supported preparation of inputs for downstream analysis.
Python
Junior AI/ML Engineer
Confidence: High Generalist
Generalist developer at a junior level focusing on web application development and basic NLP/transformers demonstrations. The strongest proven skill is building Django CRUD web applications as shown by implemented settings, views and forms code. Not evidenced are custom model training, experiment tracking, production ML serving, or performance optimization.
Model Architecture & Training
2/10
How well models are designed and trained
Minimal model-building evidence; uses pretrained Hugging Face models and pipelines with a basic QA scoring step but no custom architectures, training loops, or optimizer/schedule choices.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: AutoModelForQuestionAnswering.from_pretrained and softmax start/end scoring
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: pipeline('text-generation') and pipeline('sentiment-analysis') usage
Data Pipeline & Feature Engineering
1/10
How data is prepared for models
No engineered data pipeline or feature engineering; only tokenizer calls and basic pipeline inputs shown in notebooks.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: tokenizer = AutoTokenizer.from_pretrained(model_name)
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: batch sentiment-analysis via pipeline(task='sentiment-analysis')
Experimentation & Evaluation
2/10
How results are measured and tested
Very light evaluation practice: standalone printed QA score and example outputs but no experiment tracking, held-out splits, baselines, or reproducible experiment scaffolding.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: combined_score calculation printing QA score
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: example pipeline outputs for sentiment and generation
MLOps & Deployment
1/10
How models are shipped to production
No MLOps-specific serving or lifecycle tooling for ML models; there is standard Django deployment scaffolding for a web app but no model versioning, serving, or monitoring for ML artifacts.
Evidence
dcrm/settings.py: WSGI_APPLICATION and ALLOWED_HOSTS handling indicating web deployment configuration
manage.py: standard Django entrypoint for running the web application
Computational Efficiency
1/10
How efficiently computing resources are used
No demonstrated computational efficiency work such as quantization, batching strategies, memory profiling or GPU-optimized kernels; notebooks show nvidia-smi and use torch.no_grad but no optimization measurements.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: !nvidia-smi presence and use of torch.no_grad()
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: use of transformers pipelines for generation without efficiency tuning
Research Depth & Innovation
1/10
Depth of research and new ideas
No research-depth artifacts like paper reimplementations, custom layers, or ablation studies; work is exploratory/demo-level using existing models.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: demonstration use of pretrained models (distilbert, distilgpt2) rather than custom implementations
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: multiple pipeline examples without experimental protocol or ablation
Expertise
LLM• Junior
Industries
Artificial Intelligence• Junior
Software• Junior
Technologies
SQL
NLP
CUDA
Machine Learning
Recommendations
- Develop production-ready ML inference endpoints (simple model serving with versioning and a small test harness) to move from demo notebooks to deployable components.
- Add unit and integration tests plus CI configuration for the Django app and any model-serving endpoints to improve reliability and maintainability.
- Implement small reproducible experiments for an NLP task (train/finetune a lightweight model, log runs with W&B or MLflow, and record hyperparameters and metrics).
- Harden views by adding explicit exception handling around Record.objects.get and add input validation to avoid runtime 500s.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Junior Data Scientist
Confidence: High ML Practitioner
ML practitioner at a junior level who builds small Django web apps and experiments with pretrained Hugging Face models. The strongest proven skill is implementing CRUD web applications and user forms in Django, evident in website/forms.py and website/views.py. There is limited evidence of rigorous data science practices such as model training pipelines, validation, testing, CI, or production-grade MLOps.
Statistical Rigor
2/10
Correct use of statistics
Very limited statistical rigor; notebooks compute softmax scores and print outputs but there are no hypothesis checks, uncertainty quantification, multiple-comparison awareness, or discussion of negative results.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: question-answering section computes start/end softmax probabilities and combined_score
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: QA inference and printed score outputs
Data Wrangling & Cleaning
2/10
Preparing and cleaning data
Basic input validation and required fields appear in Django forms and tokenization is used in notebooks, but no systematic cleaning, outlier handling, dataset provenance, or leakage checks are present.
Evidence
django-CRM/website/forms.py: AddRecordForm and SignUpForm field definitions and required validators
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: tokenizer usage in question-answering example
Exploratory Analysis & Visualization
Exploring and visualizing data
Not evidenced in public code
Predictive Modeling
2/10
Building models that predict
Uses pretrained Hugging Face pipelines and runs simple inference, but there is no model training pipeline, cross-validation, calibration, error analysis, or systematic baseline-first discipline.
Evidence
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: pipeline('sentiment-analysis'), AutoModelForQuestionAnswering inference and manual scoring
huggingFace_demo/Another_copy_of_Huggingface_demo.ipynb: text-generation and sentiment-analysis pipeline usage
Business Insight & Impact
1/10
Turning analysis into business value
Implements a small CRM-style web app which is business-facing, but there is no framing of business metrics, cost-of-error reasoning, or actionable impact analysis in the code.
Evidence
django-CRM/website/views.py: CRUD endpoints, login/register flow and user-facing messages
django-CRM/website/forms.py: forms used to collect customer record fields
Reproducibility & Notebook Hygiene
2/10
Clean, repeatable analysis
Some reproducibility basics exist such as a requirements.txt and dotenv usage in settings, but notebooks are not executed / reproducible, there is no environment pinning beyond requirements, and no CI or data/versioning artifacts.
Evidence
django-CRM/requirements.txt: pinned dependency list
django-CRM/dcrm/settings.py: dotenv loading for environment configuration
Ai-fundamentals-Assessment1/Another_copy_of_Huggingface_demo.ipynb: NOTEBOOK META shows exec=none
Expertise
Data Science• Junior
Technologies
Python• since 2025 • Junior
MySQL
PostgreSQL
OpenCV
huggingface_hub
NumPy
Requests
Hugging Face
Recommendations
- Develop internal Django-based tools and CRUD dashboards where deliverables are small and well scoped.
- Prototype NLP inference services that wrap Hugging Face pipelines for QA, sentiment, or generation and focus on deployment and API integration.
- Boost data science rigor by adding training pipelines, cross-validation, error analysis, and reproducibility measures like environment pinning, seeds, and CI.
- Contribute to projects that need rapid prototypes and demos rather than large scale data engineering or production ML systems.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Intern Robotics Engineer
Confidence: Low
Early-career developer working on AI fundamentals coursework with limited public engineering artifacts. No concrete coding artifacts were available to prove a specific technical strength or to link a skill to a named file. There is no evidence of robotics-specific systems, tests, tuning data, or production-grade middleware in the provided material.
Motion Control & Kinematics
Controlling robot movement
Not evidenced in public code
Perception & Sensor Fusion
Understanding sensor data
Not evidenced in public code
ROS & Middleware
Robot software framework skills
Not evidenced in public code
Control Algorithms
Algorithms that steer robots
Not evidenced in public code
Real-time Systems
Precise real-time control
Not evidenced in public code
Simulation & HIL Testing
Testing robots in simulation
Not evidenced in public code
Recommendations
- Publish at least one small, self-authored project repository with source code, a clear README, and design notes - for robotics work prefer a single ROS/ROS2 package that includes a launch file and a short usage example.
- Deliver targeted artifacts that show engineering judgment: a sensor-fusion node with comments on timestamp alignment and covariance choices, or a motion-control node with documented PID tuning and step-response logs.
- Add tests and CI so reviewers can verify behavior: unit tests for core logic, a rosbag-based regression test, and a short recorded demo or simulation log that shows sim-to-real considerations.
- Document authorship and ownership in each file and avoid uploading only template outputs; include commit messages that explain design tradeoffs and field fixes.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories: