Shreya K Reddy
Artificial Intelligence Specialist
Node JS
C++
SQL
JavaScript
Python
C
MATLAB
Data Pipeline & Feature Engineering: 4/10
Active 5 days ago
Invite to interview
Download CVCV
Overview
Technical skills
Timeline
Roles
Overview
Applied ML engineer focused on end-to-end supervised models and LLM-backed API integration with a practical emphasis on feature engineering and application wiring. The strongest proven skill is building end-to-end ML workflows and prototypes, demonstrated by the fraud_detection.ipynb that includes dtype optimization, feature creation, SMOTE resampling and RandomForest training with evaluation. There is limited evidence of production-grade MLOps, systematic experiment tracking, automated tests or custom research implementations.
Phone
Technical skills
Languages
7
Node JS
C++
SQL
JavaScript
Python
C
MATLAB
AI/ML
12
Gemini
NumPy
Pandas
Scikit-learn
LangChain
OpenAI
Jupyter Notebook
Copilot
Machine Learning
AI/ML
RAG
Embeddings
Frontend
4
React.js
Tailwind CSS
Vite
ESLint
Databases
4
MySQL
FAISS
SQLite
Databases
DevOps
5
GitHub
Linux
Vercel
Git
Rest API
Other
5
Axios
FastAPI
Semantic Search
NLP
Semantic Search
Timeline
Dayananda Sagar University (DSU)
Bachelor's Degree •
Computer Science Engineering
Artificial Intelligence Intern
•
Junior
Codec Technologies
•
Internship
Developed backend services using Python and FastAPI to automate real-world tasks. Built RESTful APIs and integrated external APIs into application workflows. Implemented features using the OpenAI API and LangChain, including RAG-style pipelines for AI responses.
Python
FastAPI
Rest API
LangChain
OpenAI
RAG
Middle AI/ML Engineer
Confidence: Medium ML Engineer
Applied ML engineer focused on end-to-end supervised models and LLM-backed API integration with a practical emphasis on feature engineering and application wiring. The strongest proven skill is building end-to-end ML workflows and prototypes, demonstrated by the fraud_detection.ipynb that includes dtype optimization, feature creation, SMOTE resampling and RandomForest training with evaluation. There is limited evidence of production-grade MLOps, systematic experiment tracking, automated tests or custom research implementations.
Model Architecture & Training
3/10
How well models are designed and trained
Standard supervised model training using scikit-learn with basic feature engineering and imbalance handling; no custom architectures or advanced training loops.
Evidence
financial-fraud-detection-ml/fraud_detection.ipynb: RandomForestClassifier training and model.fit
financial-fraud-detection-ml/fraud_detection.ipynb: dtype mapping and feature creation (balanceOrigDiff, balanceDestDiff)
Data Pipeline & Feature Engineering
4/10
How data is prepared for models
Reasonable data preprocessing and imbalance handling including dtype optimization, explicit feature engineering, downsampling and SMOTE oversampling.
Evidence
financial-fraud-detection-ml/fraud_detection.ipynb: dtype mapping and drop of nameOrig/nameDest
financial-fraud-detection-ml/fraud_detection.ipynb: non_fraud sampling and SMOTE resampling (imblearn.over_sampling.SMOTE)
Experimentation & Evaluation
3/10
How results are measured and tested
Basic evaluation with stratified train/test split, confusion matrix, classification report and ROC AUC; no experiment tracking, ablations or systematic hyperparameter sweeps.
Evidence
financial-fraud-detection-ml/fraud_detection.ipynb: train_test_split with stratify and roc_auc_score
financial-fraud-detection-ml/fraud_detection.ipynb: confusion_matrix and classification_report printing
MLOps & Deployment
2/10
How models are shipped to production
Application-level deployment work for an LLM-backed API with persistence and health endpoints but lacking model serving, versioning, monitoring, authentication and deployment automation.
Evidence
ai-healthcare-assistant/backend/main.py: FastAPI app with /chat and /chat-history endpoints and MongoDB persistence
ai-healthcare-assistant/backend/main.py: /health endpoint and CORSMiddleware configuration
Computational Efficiency
2/10
How efficiently computing resources are used
Minimal efficiency work: dtype casting for memory and use of n_jobs for parallel trees; no GPU, quantization, profiling or batching optimizations.
Evidence
financial-fraud-detection-ml/fraud_detection.ipynb: dtype mapping for memory optimization
financial-fraud-detection-ml/fraud_detection.ipynb: RandomForestClassifier(n_jobs=-1)
Research Depth & Innovation
1/10
Depth of research and new ideas
No evidence of research-grade depth, novel algorithms, custom layers or reproduced paper implementations.
Evidence
financial-fraud-detection-ml/fraud_detection.ipynb: standard scikit-learn pipeline without custom architectures
Expertise
Finance & FinTech AI• Middle
Medical AI & Healthcare• Middle
Industries
Financial Services• Middle
Health Care• Middle
Technologies
Databases
AI/ML
Python• since 2026 • Junior
SQL
C++
MATLAB• since 2026 • Junior
MySQL
Copilot
LangChain• since 2026
FAISS
Jupyter Notebook
Vercel
Embeddings
Scikit-learn• since 2026
NLP• since 2026
Pandas• since 2026
NumPy• since 2026
Git• since 2026
SQLite
Gemini
RAG• since 2026
Semantic Search
OpenAI• since 2026
GitHub• since 2026
Semantic Search
Linux
Machine Learning
Computer Vision• mentioned only
Machine Learning• mentioned only
Vision• mentioned only
Recommendations
- Develop prototype fraud scoring and transaction monitoring pipelines that include data validation, feature stores and model retraining automation.
- Implement LLM-backed conversational services with hardened guardrails, input validation, authentication and monitoring for safety-critical domains like healthcare.
- Package models and APIs for production with model versioning, CI/CD, automated tests and basic observability (metrics/logs/alerts).
- Improve experimentation practices by adding reproducible experiment tracking, hyperparameter searches and ablation studies.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Junior Frontend Developer
Confidence: Medium UI Engineer
Junior frontend engineer specializing in React and UI styling for Vite-based projects. The strongest proven skill is CSS theming and responsive UI work as demonstrated in src/index.css and src/App.css across multiple frontend folders. There is little public evidence of complex state management, async data flow discipline, test coverage, or backend and performance engineering.
UI Component Architecture
2/10
How interface parts are built
Minimal component architecture; artifacts are largely CSS and template scaffolding with no evidence of custom React component design or composition patterns.
Evidence
ai_research_assistant/frontend/src/App.css
ai-sql-query-generator/frontend/frontend/src/App.css
Responsive & Cross-browser
3/10
Works on all screens and browsers
Responsive and theming work is present using media queries and prefers-color-scheme, but there is no advanced cross-browser feature detection or RTL/i18n scaffolding.
Evidence
ai_research_assistant/frontend/src/index.css
ai_research_assistant/frontend/src/App.css
Performance Optimization
1/10
Speed of the interface
No measured or explicit performance engineering; standard Vite setup exists but no bundle analysis, code-splitting, virtualization or documented optimizations.
Evidence
ai_research_assistant/frontend/vite.config.js
Accessibility & Semantics
3/10
Usable for everyone
Basic accessibility touches (focus-visible outlines, semantic HTML container) are present, but there is no evidence of ARIA on custom widgets, keyboard-management, or automated a11y checks in CI.
Evidence
ai_research_assistant/frontend/src/App.css
ai_research_assistant/frontend/index.html
State Management & Data Flow
1/10
Managing data in the app
No structured state-management or async discipline is shown; only dependency hints for networking exist without implementation of cancellation, optimistic updates or state machines.
Evidence
rag-ai-assistant/frontend/package.json
UX & Visual Polish
3/10
Look and feel quality
Good visual polish in CSS variables, theming and responsive layout, but missing skeletons, comprehensive loading/error/empty states and advanced UX patterns.
Evidence
ai_research_assistant/frontend/src/index.css
ai_research_assistant/frontend/src/App.css
Expertise
React• Junior
Frontend Architecture & Build Tools• Junior
HTML & CSS• Junior
Industries
Artificial Intelligence• Junior
Data & Analytics• Junior
Technologies
JavaScript
Node JS• Junior
Tailwind CSS
React.js
Vite
Axios
ESLint
Recommendations
- Use for frontend UI theming, responsive layout work and converting visual templates into reusable React components.
- Task with setting up and hardening build and lint tooling (Vite, ESLint, Tailwind) and migrating styles into a component-based design system.
- Assign to improve basic accessibility (ARIA, keyboard behaviour) and to implement robust loading/error/empty states and simple async flows with request cancellation.
- Avoid assigning advanced state-heavy features, critical performance tuning, or backend integrations without senior oversight.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Junior Game Developer
Confidence: Medium Generalist
Junior frontend developer specializing in React and modern JS build tooling with a focus on correct project configuration. The strongest proven skill is project toolchain configuration as shown in frontend/vite.config.js and frontend/eslint.config.js. There is no public evidence of backend logic, AI model implementation, game systems, or performance-focused engineering in the analyzed human-authored files.
Gameplay Systems & Mechanics
How game logic works
Not evidenced in public code
Graphics & Rendering
Drawing game visuals
Not evidenced in public code
Physics & Math
Game physics and math
Not evidenced in public code
Engine Proficiency
2/10
Skill with the game engine
Basic frontend toolchain and build configuration for React+Vite and ESLint is present, showing minimal but correct use of modern JS tooling beyond an untouched template.
Performance & Frame Budget
Keeping the game smooth
Not evidenced in public code
Content Pipeline & Tooling
Tools for game content
Not evidenced in public code
Industries
Artificial Intelligence• Junior
Recommendations
- Use for frontend engineering tasks: React component work, Vite build setup, and ESLint/tooling improvements.
- Implement small full-stack prototypes by adding clear backend endpoints and end-to-end examples to demonstrate integration skills.
- Expand to measurable engineering work such as performance profiling, automated tests, or CI to raise evidence of systems-level competence.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories: