Godfrey Purification
Python Developer
5+ years exp
JavaScript
Node JS
C++
TypeScript
Python
C
Data Pipeline & Feature Engineering: 4/10
Active 1 day ago
Invite to interview
Download CVCV
Overview
Technical skills
Projects
Timeline
Roles
Overview
Backend API engineer at a Middle level focused on building simple, well-structured Node.js REST services for a child-focused learning application. The strongest proven skill is service-layer and controller implementation as shown in src/services/learning.service.js and src/controllers/progress.controller.js. There is limited evidence of database schema evolution, transaction handling, production-grade observability, load scaling, or automated tests.
Phone
Technical skills
Languages
6
JavaScript
Node JS
C++
TypeScript
Python
C
Python
3
Django
FastAPI
Flask
Node JS
3
Express
Axios
Mongoose
DevOps
3
Rest API
Git
CI/CD
Other
17
React.js
Vue.js
NumPy
Helmet
Keras
LLM
Oracle
Redux
GitHub
Vite
Dotenv
ESLint
GraphQL
AI/ML
Deep Learning
LLM Apps
Machine Learning
Projects
Timeline
Freelancer
•
Middle
Upwork
•
Freelance
Develops and maintains responsive full-stack web applications based on client requirements. Builds frontend components with React and Vue and implements backend services and REST APIs using Node.js, Express, Flask, Django, and FastAPI. Optimizes performance, scalability, and security, implements database solutions with Oracle, and automates workflows using n8n and API integrations. Uses Git with CI/CD workflows for delivery and collaboration.
Node JS
Express
Flask
Django
FastAPI
React.js
Vue.js
JavaScript
Rest API
n8n
Git
CI/CD
Asian University of Bangladesh
Master's Degree •
Marketing
National University, Bangladesh
Master's Degree •
Business Administration
Middle AI/ML Engineer
Confidence: Medium ML Engineer
Applied ML engineer producing small end-to-end prototypes in computer-vision and simple reinforcement-learning domains at a middle level. The strongest proven skill is input preprocessing for handwritten digit recognition, demonstrated by the preprocess and center_by_mass implementations in digit_recognization_keras/model_with_keras.py. There is little evidence of production MLOps, systematic evaluation, experiment tracking, or research-grade model development in public code.
Model Architecture & Training
3/10
How well models are designed and trained
Basic model building and training with Keras and a simple RL value update; correct but elementary architectures and training practice.
Evidence
digit_recognization_keras/model_with_keras.py: Sequential model definition and model.fit(...)
deeplearning_tic_tac_too/tic_tac_to.ipynb: Player.feedReward method implementing value updates
Data Pipeline & Feature Engineering
4/10
How data is prepared for models
Concrete, correct preprocessing for handwritten input (bbox, resize, center-of-mass) and normalization suitable for MNIST-style inference.
Evidence
digit_recognization_keras/model_with_keras.py: preprocess(pil_img) implementation
digit_recognization_keras/model_with_keras.py: center_by_mass(arr) implementation
Experimentation & Evaluation
1/10
How results are measured and tested
Minimal experimentation and evaluation practice; single training runs with no held-out evaluation, no experiment tracking or reproducible configs.
Evidence
digit_recognization_keras/model_with_keras.py: model.fit(x_train, y_train, epochs=3, batch_size=64)
deeplearning_tic_tac_too/tic_tac_to.ipynb: print('Rounds ...') logging and no structured eval metrics
MLOps & Deployment
1/10
How models are shipped to production
Local demo UIs present but no production serving, versioning, monitoring, or model packaging for deployment.
Evidence
digit_recognization_keras/model_with_keras.py: App class using tkinter and local GUI loop
AI-CHATBOT-PYTHON/app.py: gradio.Interface(...).launch() used for demo
Computational Efficiency
1/10
How efficiently computing resources are used
No targeted computational-efficiency work; uses default training settings and no profiling or advanced acceleration.
Evidence
digit_recognization_keras/model_with_keras.py: model.fit(..., batch_size=64)
requirements.txt: standard ML libs listed but no GPU/quantization/efficient-inference artifacts in human-authored code
Research Depth & Innovation
1/10
Depth of research and new ideas
No research-grade contributions, no custom layers, novel algorithms or reproduced papers; implementations are instructional/demo quality.
Evidence
digit_recognization_keras/model_with_keras.py: simple Keras Sequential MLP (Flatten / Dense)
deeplearning_tic_tac_to/tic_tac_to.ipynb: classic table/dictionary-based RL agent
Expertise
Conversational AI & Chatbots• Junior
Technologies
AI/ML
Deep Learning
LLM Apps
C++
Oracle
NumPy
Keras
LLM• since 2026
Machine Learning• since 2026
Recommendations
- Build a reproducible experiment pipeline: add train/validation splits, evaluation scripts, metrics logging and experiment tracking (W&B or MLflow) for the MNIST work.
- Harden the chatbot prototype into a production-capable service: add retry/backoff, input sanitization, structured logging, and a simple CI/CD pipeline with tests.
- Improve MLOps and reproducibility: package models (saved checkpoints), add inference wrappers (REST or TorchServe/BentoML), and include unit tests and a minimal deployment manifest.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Middle Backend Developer
Confidence: Medium API Engineer
Backend API engineer at a Middle level focused on building simple, well-structured Node.js REST services for a child-focused learning application. The strongest proven skill is service-layer and controller implementation as shown in src/services/learning.service.js and src/controllers/progress.controller.js. There is limited evidence of database schema evolution, transaction handling, production-grade observability, load scaling, or automated tests.
API Design
3/10
How well APIs are designed
Basic REST API design is present with a simple client contract and consistent error handling patterns but lacks versioning, idempotency, pagination or formal error schemas.
Evidence
BRIMSEL/kids-english-ui-frontend/src/services/api.js
BRIMSEL/kids-english-backend/src/controllers/progress.controller.js
BRIMSEL/kids-english-backend/src/utils/AppError.js
Data Layer & Database
2/10
Working with databases
Uses Mongoose for persistence with simple find/save usage but no migration history, transaction management, or advanced query tuning is evident.
Evidence
BRIMSEL/kids-english-backend/package.json (mongoose dependency)
BRIMSEL/kids-english-backend/src/controllers/progress.controller.js
Scalability & Performance
1/10
Handling load and speed
Almost no scalability or performance work visible; dependencies indicate awareness (rate-limit, helmet) but no caching, queuing, or measured optimizations are implemented in code provided.
Evidence
BRIMSEL/kids-english-backend/package.json (express-rate-limit, helmet)
System Architecture
3/10
Overall system structure
Reasonable module separation (services, controllers, utils) for a single-node API service but no evidence of deliberate distributed architecture, service contracts or graceful degradation strategies.
Evidence
BRIMSEL/kids-english-backend/src/services/learning.service.js
BRIMSEL/kids-english-backend/src/services/chatbot.service.js
BRIMSEL/kids-english-backend/src/controllers/progress.controller.js
Security & Auth
2/10
Protecting data and access
Some basic security awareness via dependencies and input normalization, but no auth/authz flows, secret management, or strong boundary protection are implemented in the examined code.
Evidence
BRIMSEL/kids-english-backend/package.json (helmet, dotenv)
BRIMSEL/kids-english-backend/src/services/learning.service.js
BRIMSEL/kids-english-backend/src/utils/AppError.js
Reliability & Observability
3/10
Stability and monitoring
Basic reliability patterns exist: a custom AppError and asyncHandler are used to centralize error handling, but there is no structured logging, timeouts, retry/backoff, or metrics visible.
Evidence
BRIMSEL/kids-english-backend/src/controllers/progress.controller.js
BRIMSEL/kids-english-backend/src/utils/asyncHandler.js
BRIMSEL/kids-english-backend/src/utils/AppError.js
Expertise
Node.js• Middle
Microservices & API Architecture• Middle
Industries
Education• Middle
Technologies
Python
Rest API• since 2021
GraphQL
Flask• since 2021
FastAPI• since 2021
Django• since 2021
Helmet
Dotenv
Recommendations
- Harden the backend for production: add authentication, secrets management, structured logging, and metrics.
- Add database migration support and explicit transaction boundaries for data integrity and schema evolution.
- Introduce automated tests (unit and integration) and add CI checks to prevent regressions.
- Fix logic bugs and improve input contracts - e.g., ensure service return shapes are consistent and documented (teach() should return any flags consumed by callers).
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Middle Frontend Developer
Confidence: Medium Fullstack
A middle-level fullstack developer building interactive React and React Native applications with a focus on small-scale educational experiences. The strongest proven skill is designing and implementing the learning and conversational backend flow, evidenced by learning.service.js and agent.service.js which implement question selection, answer checking, mode detection and progress updates. Public code shows visual polish and basic responsive design but lacks test coverage, CI, documented accessibility work and measurable performance optimizations.
UI Component Architecture
3/10
How interface parts are built
Basic UI styling, tokens and some shared color constants exist, but there is no evidence of a deliberate component library, composition patterns, or non-trivial custom hooks driving component boundaries.
Evidence
BRIMSEL/kids-english-ui-frontend/src/App.css: global layout, message and chat input styles
BRIMSEL/kids-english-ui-frontend/src/index.css: design tokens and responsive typography
myapp_react_native/constants/Colors.js: shared color tokens
Responsive & Cross-browser
4/10
Works on all screens and browsers
Responsive layout and dark-mode support are present via media queries and prefers-color-scheme, but there is no advanced feature-detection, RTL/i18n scaffolding or container-query usage.
Evidence
BRIMSEL/kids-english-ui-frontend/src/App.css: @media (max-width: 768px) layout adjustments
BRIMSEL/kids-english-ui-frontend/src/index.css: @media (max-width: 1024px) and prefers-color-scheme: dark
Performance Optimization
2/10
Speed of the interface
No measured performance artifacts or explicit optimizations such as virtualization, bundle splits, or before/after metrics; only basic build tooling usage is visible.
Evidence
BRIMSEL/kids-english-ui-frontend/vite.config.js: Vite + plugin-react usage
BRIMSEL/kids-english-ui-frontend/package.json: app dependencies and build scripts
Accessibility & Semantics
2/10
Usable for everyone
Minor a11y-conscious styles (focus states, prefers-color-scheme) exist but there is no explicit ARIA handling, keyboard management, automated a11y checks or accessibility CI shown.
Evidence
BRIMSEL/kids-english-ui-frontend/src/App.css: .chat-input-form input:focus
BRIMSEL/kids-english-ui-frontend/src/index.css: color-scheme and dark-mode token adjustments
State Management & Data Flow
4/10
Managing data in the app
Server and domain state handling is present with clear service boundaries and simple state transitions (learning flow, mode detection, progress resets), but there is no evidence of advanced client-server state management patterns like optimistic updates with rollback, request cancellation, or documented cache invalidation.
Evidence
BRIMSEL/kids-english-backend/src/services/learning.service.js: teach, checkAnswer, question selection and child state updates
BRIMSEL/kids-english-backend/src/services/agent.service.js: detectMode and handleMessage integrating learning.service
BRIMSEL/kids-english-backend/src/controllers/progress.controller.js: progress read/reset using asyncHandler and AppError
UX & Visual Polish
4/10
Look and feel quality
Good visual polish for a small app - animations, gradients, disabled states and a typing indicator are implemented, and there are responsive considerations, but UX fallbacks, extensive empty/error states and undo patterns are limited.
Evidence
BRIMSEL/kids-english-ui-frontend/src/App.css: typing-indicator animation and chat UX styles
BRIMSEL/kids-english-ui-frontend/src/App.css: disabled button and input focus states
BRIMSEL/kids-english-ui-frontend/src/index.css: design tokens and layout rules
Expertise
React• Middle
Industries
Education• Middle
Technologies
JavaScript• since 2021 • Senior
TypeScript
Node JS• since 2021 • Senior
Vue.js• since 2021
Redux
Express• since 2021
CI/CD• since 2021
Git• since 2021
React.js• since 2021
Vite
Mongoose
Axios
ESLint
GitHub
Recommendations
- Develop cross-platform React Native mobile features and reusable UI components for education-focused apps, pairing Expo Router routes with well-scoped components.
- Implement and extend server-side learning flows and conversational APIs using Node.js and Express with Mongoose-backed models and robust error handling.
- Improve engineering maturity by adding automated tests, CI pipelines, accessibility audits and measured performance profiling (LCP/INP or bundle analysis).
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories: