Aidar Ibatullin
Python
Node JS
JavaScript
Active 17 days ago
+7 (937) 1515153 Invite to interview
Message
Download CVCV
Overview
Technical skills
Roles
Overview
A pragmatic backend engineer at a solid middle level who builds customer-facing APIs and integrations end-to-end. The strongest proven skill is building messaging and auth flows with persistence and security measures, demonstrated by the Telegram bot (skysamurai/kolbot/bot/bot.py) and the FastAPI auth service (skysamurai/kolbot/example_webapp_auth/backend.py). There is limited evidence of large-scale distributed architecture, formal observability, or extensive schema migration history in the human-authored code.
Technical skills
Python• Middle
Node JS• Middle
JavaScript• 3y+
Python
FastAPI
HTTPX
Requests
Uvicorn
Pydantic
Node JS
Bcrypt
Databases
SQLite
Supabase
Middle AI/ML Engineer
Confidence: Medium Generalist
A pragmatic generalist developer at a solid junior-to-middle level who builds production web backends, Telegram chatbots and automation glue. The strongest proven skill is backend+integration engineering - evidenced by the Telegram bot and SQLite-backed receipt/purchase pipeline (kolbot/bot/bot.py and kolbot/bot/db.py). There is no evidence of custom model training, rigorous ML evaluation, GPU/quantization work, or advanced MLOps/model serving in the human-authored files.
Model Architecture & Training
1/10
How well models are designed and trained
Minimal ML model work - only thin ML/AI API integration for receipt parsing; no model training, custom loss, or architecture design.
Evidence
kolbot/bot/receipt_parser.py: functions parse_receipt, _try_deepseek_with_url (DeepSeek API integration for receipt parsing)
Data Pipeline & Feature Engineering
3/10
How data is prepared for models
Solid pragmatic data handling and schema design - clear SQLite schema, migrations/seeding and ETL-like receipt -> purchases flow plus workflow automation for batch updates.
Evidence
kolbot/bot/db.py: SCHEMA definition and functions create_receipt, create_purchase, get_user_purchases
kolbot/workflow_v2/build_v2_part5.py: Supabase REST calls and cron/workflow code that performs batch updates and event recording
Experimentation & Evaluation
How results are measured and tested
Not evidenced in public code
MLOps & Deployment
1/10
How models are shipped to production
Basic deployment and integration engineering - web endpoints, webhook handlers and uvicorn entrypoints exist, but no model-serving, versioning, or drift-monitoring MLOps features.
Evidence
kolbot/example_webapp_auth/backend.py: FastAPI endpoints, uvicorn.run, webhooks and session handling
kolbot/workflow_v2/build_v2_part5.py: webhook/payment and cron/auto-approve workers wired to Supabase REST endpoints
Computational Efficiency
1/10
How efficiently computing resources are used
Little to no computational efficiency engineering - no GPU/quantization/distributed training or profiling work in the human-authored files.
Evidence
kolbot/workflow_v2/build_v2_part5.py: batched REST/cron operations and session reuse (requests.Session) - basic efficiency considerations
Research Depth & Innovation
1/10
Depth of research and new ideas
No research-depth artifacts - no paper implementations or novel algorithms; a practical prompt/inference integration appears in receipt parsing only.
Evidence
kolbot/bot/receipt_parser.py: code paths calling external AI (DeepSeek) and parsing results
Expertise
Conversational AI & Chatbots• Middle
Industries
Commerce• Middle
Sales & Marketing• Middle
Technologies
Pydantic
Uvicorn
Recommendations
- Build and maintain conversational bots and web backends that integrate third-party ML APIs (Telegram bot flows, receipt parsing pipelines).
- Implement payment/workflow automations and Supabase or SQLite-backed server components (webhooks, cron workers, n8n integrations).
- Prototype integrations between hosted AI/ML services and product logic (structured parsers, post-processing, idempotent workflows).
- Own small-to-medium backend features: auth, session handling, DB migrations and admin tooling for consumer-facing services.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Middle Backend Developer
Confidence: High API Engineer
A pragmatic backend engineer at a solid middle level who builds customer-facing APIs and integrations end-to-end. The strongest proven skill is building messaging and auth flows with persistence and security measures, demonstrated by the Telegram bot (skysamurai/kolbot/bot/bot.py) and the FastAPI auth service (skysamurai/kolbot/example_webapp_auth/backend.py). There is limited evidence of large-scale distributed architecture, formal observability, or extensive schema migration history in the human-authored code.
API Design
5/10
How well APIs are designed
Solid REST API endpoints with JWT sessions, explicit error handling and idempotency checks for webhooks; lacks a formal versioning strategy and consistent pagination conventions across services.
Data Layer & Database
5/10
Working with databases
Thoughtful relational schema design with indexes, foreign keys and migration-friendly init code; uses PRAGMA and unique constraints to protect integrity, but no long migration chain is present in the human-authored files.
Scalability & Performance
3/10
Handling load and speed
Some operational choices for reliability and modest concurrency (WAL, thread-local connection, HTTP retry loops) but no explicit connection pooling, caching or queue-based decoupling for scale.
System Architecture
4/10
Overall system structure
Clear separation between bot logic, persistence and an auth API plus n8n workflow glue demonstrates deliberate module boundaries and integration planning; not a distributed microservice architecture by itself.
Security & Auth
6/10
Protecting data and access
Good practical security awareness - HMAC validation for Telegram initData, bcrypt for passwords, JWT sessions, explicit internal-secret checks and idempotency for payments; however some deployment scripts contain hardcoded credentials which is a risk.
Evidence
skysamurai/kolbot/example_webapp_auth/backend.py: validate_telegram_init_data uses HMAC with hmac.compare_digest; bcrypt.hashpw and JWT token creation; INTERNAL_SECRET checks for confirm-phone and link-by-code
skysamurai/kolbot/bot/db.py: payments table includes idempotency_key UNIQUE to prevent duplicate payment processing
Reliability & Observability
4/10
Stability and monitoring
Practical reliability patterns are present - retries, deduplication, persisted offsets and WAL - but limited structured observability, metrics, or graceful shutdown handling in the human-authored code.
Expertise
Python• Middle
Messaging & Real-time• Middle
Microservices & API Architecture• Middle
Databases & Vector Storage• Middle
Industries
Commerce• Middle
Technologies
Python• Middle
Node JS• Middle
Supabase
FastAPI
Bcrypt
SQLite
HTTPX
Requests
Recommendations
- Use to implement and maintain customer-facing backend APIs (auth, payments, referral flows) and Telegram integrations.
- Use to build and harden small-to-medium services that require reliable SQLite/Postgres schemas, idempotent webhooks and transactional flows.
- Use to implement workflow automation and glue (n8n / REST) between services, including payment/webhook handling and admin tooling.
- Avoid assigning as sole architect for high-scale distributed systems or low-latency trading systems without additional senior design oversight.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories:
Intern Frontend Developer
Confidence: Medium UI Engineer
A front-end focused developer at a Junior level who delivers polished static landing pages with well-considered visual design. The strongest proven skill is crafting pixel-accurate, responsive marketing UIs as evidenced by the hero, programs and modal implementations in the HTML/CSS files. The codebase shows minimal application-state complexity, no advanced accessibility focus management, and no evidence of component architecture or measured performance work.
UI Component Architecture
2/10
How interface parts are built
Small, hand-coded page structure with repeated card patterns but no componentization or custom hook logic; CSS is inline and layout is page-scoped rather than component-driven.
Responsive & Cross-browser
4/10
Works on all screens and browsers
Responsive layout via media queries and fluid grid adjustments; straightforward approach that addresses common breakpoints but lacks advanced techniques (container queries, RTL or explicit feature-detection).
Performance Optimization
3/10
Speed of the interface
Basic performance considerations present (lazy-loaded images, minimal JS), but no measured optimizations, code-splitting, or virtualization for larger data sets.
Accessibility & Semantics
3/10
Usable for everyone
Some a11y-aware choices (use of <button>, aria-expanded, Escape key handling) but missing focus management for modal flows, no focus trap/return, and limited ARIA on custom widgets.
State Management & Data Flow
2/10
Managing data in the app
UI state is managed with simple DOM toggles and show/hide logic; there is no async state discipline, request cancellation, optimistic updates, or form validation beyond required attributes.
UX & Visual Polish
5/10
Look and feel quality
High visual polish for a static landing - consistent tokens, well-crafted buttons, spacing, and transitions; UX covers common happy paths with success state and graceful mobile layout.
Expertise
HTML & CSS• Intern
Industries
Lifestyle• Intern
Technologies
JavaScript• 3y+
Recommendations
- Own and extend marketing/landing pages and UI kits - implement accessible modals (focus trap, return focus) and improved form validation with progressive enhancement.
- Refactor repeated UI into reusable components (Web Components or a small framework) and extract design tokens into a single stylesheet for maintainability.
- Practice accessible-interaction patterns and automated a11y checks (axe or eslint-plugin-jsx-a11y) to raise accessibility from basic to production-grade.
Repositories
The developer's experience in this domain has been verified based on AI analysis of the following repositories: