Vivekanand Choudhary - Software Developer

Overview

Senior backend engineer focused on building production-grade distributed Python services that combine event-driven automation and LLM-backed features. The strongest proven skill is reliable event ingestion and execution with exactly-once semantics and DLQ/retry handling as implemented in app/ingestion/webhooks.py and app/worker/tasks.py. There is limited public evidence of large-scale ops (Kubernetes manifests, multi-stage CI/CD, or large-scale load-testing reports).

Phone

Timeline

Software Engineer · Middle

Proficonlabs Pvt Ltd. (Simpo.ai) · Full-Time

Nov 2024 to Present 2 Years Hyderabad In office

Developed backend services and custom processing operators using Java and Spring Boot as part of microservices pipelines. Built tenant-isolation improvements to protect high-throughput customers from impacting others. Integrated an industry keyword pipeline with competition-level categorization and implemented services for file storage and UI-triggered updates. Mentored and onboarded interns to contribute independently while following engineering practices.

Java

Spring Boot

Software Intern · Junior

Proficonlabs Pvt Ltd. (Simpo.ai) · Full-Time

Aug 2024 to Oct 2024 2 Months Hyderabad In office

Assisted with engineering work supporting ongoing production systems. Contributed to development and maintenance tasks aimed at reducing issues and keeping services running reliably.

Projects

Event-Triggered Automation Agent

Developed a backend automation system that reacts to events such as payments and third-party updates, determines the required actions, and executes them automatically. Implemented retry handling and safe storage for repeatedly failing tasks to avoid lost work. Added exactly-once style workflow processing and real-time progress visibility for users via a GraphQL interface.

ARX-24+ Model-AI-Chat-Platform

Built a multi-model LLM chat platform that generates grounded answers using private documents to reduce hallucinations. Implemented a hybrid RAG pipeline combining semantic caching, vector search, and re-ranking, with low-latency streaming. Added a zero-restart model registry so new models can be deployed without code changes and exposed RAG actions as structured tool calls.

Technical skills

Languages

5

TypeScript

Java

Python

SQL

JavaScript

Python

7

Celery

FastAPI

SQLAlchemy

Pydantic

Asyncio

Alembic

Uvicorn

Java

3

Spring Boot

Hibernate

Spring Security

Databases

5

PostgreSQL

Redis

MongoDB

MySQL

Cassandra

AI/ML

4

OpenAI

Anthropic

LLM

LangChain

DevOps

11

OpenTelemetry

AWS

Azure

Amazon S3

Kubernetes

Amazon EC2

AWS Lambda

GitHub

Docker

Git

Rest API

Other

17

React.js

Tailwind CSS

Next.js

Recharts

Vite

PostCSS

Model Context Protocol

AI Agents

CI/CD

Jenkins

gRPC

GraphQL

RAG

autoprefixer

Function Calling

IAM

Hallucination

Senior AI/ML Engineer Confidence: Medium LLM Engineer

LLM-focused engineer at a senior level with strong systems and inference-integration skills for multi-model chat and RAG platforms. The strongest proven skill is building production LLM inference and RAG pipelines as shown by the streaming chat router, model registry and vectorstore integration (app/services/llm.py, app/services/registry.py, app/services/vectorstore.py). There is little to no evidence of custom model training, research implementations or experiment tracking in public code.

Model Architecture & Training

2/10

How well models are designed and trained

Little to no custom model architecture or training code; mainly inference, routing and integration engineering around hosted LLMs.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/llm.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/registry.py

Data Pipeline & Feature Engineering

4/10

How data is prepared for models

Clear, production-minded document ingestion and embedding pipeline for RAG with chunking, parsing and vectorstore integration.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/ingest.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/vectorstore.py

Event-Triggered-Automation-Agent/app/ingestion/webhooks.py

Experimentation & Evaluation

1/10

How results are measured and tested

Basic automated tests and focused unit tests exist but no experiment tracking, sweep configurations or ML evaluation pipelines.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/tests/test_retrieval.py

ARX-24-Model-AI-Chat-Platform/arx/backend/tests/test_fallback.py

MLOps & Deployment

5/10

How models are shipped to production

Strong production inference and MLOps engineering: model registry, routing/failover, SSE streaming, quotas, metrics and tracing are implemented.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/registry.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/routers/chat.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/quota.py

Computational Efficiency

2/10

How efficiently computing resources are used

Reasonable async and concurrency patterns are used and embedding is moved off-thread, but no GPU optimizations, quantization or explicit profiling artifacts are present.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/vectorstore.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/ingest.py

Research Depth & Innovation

1/10

Depth of research and new ideas

No original research, custom layers or SOTA reproductions; the work is engineering-focused rather than research-focused.

Evidence

Event-Triggered-Automation-Agent/app/agent/decision.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/llm.py

Expertise

LLM· Senior

RAG· Senior

Industries

Artificial Intelligence· Senior

Software· Senior

Technologies

Python· Senior

SQL

MySQL

MongoDB

PostgreSQL

Redis

LangChain

Cassandra

Model Context Protocol

OpenTelemetry

Function Calling

AI Agents

Azure

CI/CD

Jenkins

Git

AWS

Docker

Kubernetes

LLM

RAG

Pydantic

Alembic

AWS Lambda

Amazon EC2

Hallucination

OpenAI

Anthropic

GitHub

Amazon S3

IAM

Recommendations

Develop multi-model inference platforms and RAG systems with streaming and failover logic.
Implement production event-driven automation and orchestration using Celery, gRPC and reliable idempotent webhooks.
Build model registry, routing, cost-estimation and quota/billing components for hosted LLM integrations.
Extend test coverage with reproducible ML evaluation pipelines and add experiment tracking for any future fine-tuning work.

Repositories

The developer's experience in this domain has been verified based on AI analysis of the following repositories:

ARX-24-Model-AI-Chat-Platform Event-Triggered-Automation-Agent

Senior Backend Developer Confidence: High Distributed Systems

Senior backend engineer focused on building production-grade distributed Python services that combine event-driven automation and LLM-backed features. The strongest proven skill is reliable event ingestion and execution with exactly-once semantics and DLQ/retry handling as implemented in app/ingestion/webhooks.py and app/worker/tasks.py. There is limited public evidence of large-scale ops (Kubernetes manifests, multi-stage CI/CD, or large-scale load-testing reports).

API Design

6/10

How well APIs are designed

APIs are well designed for real use: SSE streaming, origin guard, explicit idempotent webhook ingestion, and consistent error models are present.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/app/routers/chat.py

Event-Triggered-Automation-Agent/app/ingestion/webhooks.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/security.py

Data Layer & Database

7/10

Working with databases

Data layer shows deliberate schema design and evolution, DB constraints and indices, SQLAlchemy models with relationships, and migration history via Alembic.

Evidence

Event-Triggered-Automation-Agent/alembic/versions/001_initial_schema.py

Event-Triggered-Automation-Agent/app/models.py

Event-Triggered-Automation-Agent/app/graphql_api/schema.py

Scalability & Performance

6/10

Handling load and speed

Scalability practices are evident: Celery workers with retries and DLQ, quota backends with Redis fallback, registry caching with TTL, and vectorstore embed offload to threads.

Evidence

Event-Triggered-Automation-Agent/app/worker/tasks.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/quota.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/vectorstore.py

System Architecture

6/10

Overall system structure

System decomposition is deliberate with separate ingestion, worker, gRPC, GraphQL and REST layers, and centralized configuration via pydantic settings.

Evidence

Event-Triggered-Automation-Agent/app/graphql_api/main.py

Event-Triggered-Automation-Agent/app/grpc_server/server.py

ARX-24-Model-AI-Chat-Platform/arx/backend/app/main.py

Security & Auth

7/10

Protecting data and access

Security-focused decisions are visible: password hashing, JWT sessions with revocation, origin/internal-key checks, and webhook signature verification.

Evidence

ARX-24-Model-AI-Chat-Platform/arx/backend/app/services/security.py

Event-Triggered-Automation-Agent/app/ingestion/signatures.py

Event-Triggered-Automation-Agent/app/ingestion/webhooks.py

Reliability & Observability

7/10

Stability and monitoring

Reliability and observability are well addressed through retries with backoff, dead-letter capture, idempotency, Prometheus metrics and OpenTelemetry instrumentation, plus unit tests for critical paths.

Evidence

Event-Triggered-Automation-Agent/app/worker/tasks.py

Event-Triggered-Automation-Agent/app/ingestion/idempotency.py

Event-Triggered-Automation-Agent/app/observability.py

Expertise

Backend AI & LLM· Senior

Databases & Vector Storage· Senior

Messaging & Real-time· Senior

Microservices & API Architecture· Senior

Python· Senior

System Architecture· Senior

Industries

Artificial Intelligence· Senior

Technologies

Java· since 2024 · Middle

Rest API

gRPC

GraphQL

Spring Boot· since 2024

SQLAlchemy

Hibernate

FastAPI

Asyncio

Celery

Spring Security

Uvicorn

Recommendations

Develop event-driven backend systems that require robust idempotency, DLQ handling and observability (webhooks, task queues, retries).
Build or extend LLM-backed RAG chat services and vector search integrations, including SSE streaming and model routing logic.
Implement backend integrations that combine gRPC workers, Celery tasks and GraphQL or REST frontends with strong telemetry.
Design database schema evolution and migration strategies for systems using Postgres+JSONB and secondary stores like Chroma.

Repositories

The developer's experience in this domain has been verified based on AI analysis of the following repositories:

ARX-24-Model-AI-Chat-Platform Event-Triggered-Automation-Agent Interview_Prep-Tracker

Middle Frontend Developer Confidence: Medium UI Engineer

Frontend-focused UI engineer at a middle level with a clear strength in building interactive, accessible and performance-conscious web interfaces. The strongest proven skill is accessible, high-polish UI and runtime performance work, demonstrated in the interactive portfolio page where dialog focus management, skip-links, aria-live toast, IntersectionObserver reveals and requestAnimationFrame scroll-throttling are implemented. There is limited evidence of automated tests, advanced server-state patterns like optimistic updates with rollbacks, or large-scale React component library design in public code.

UI Component Architecture

3/10

How interface parts are built

Basic component patterns with isolated custom hooks and tokenized styling; mostly page-level HTML rather than a component library.

Evidence

Interview_Prep-Tracker/src/hooks/usePrepData.ts: custom React hook managing server CRUD and demo fallback

Interview_Prep-Tracker/tailwind.config.js: design tokens and theme extension

portfolio/index.html: modular JS functions (openProject, showToast) separating behaviour from markup

Responsive & Cross-browser

6/10

Works on all screens and browsers

Thoughtful responsive layout and dark-mode theming with explicit media queries and reduced-motion support; good cross-size handling but no advanced feature-detection or container queries.

Evidence

portfolio/index.html: responsive CSS media queries and layout adjustments for many breakpoints

Interview_Prep-Tracker/src/index.css: CSS variables for light/dark themes and layout baseline

Performance Optimization

6/10

Speed of the interface

Measured runtime optimizations and animation discipline using requestAnimationFrame, passive scroll listeners and IntersectionObserver to limit work on scroll and reveal elements.

Evidence

portfolio/index.html: scroll handling using requestAnimationFrame with passive listener and scroll-linked progress bar

portfolio/index.html: IntersectionObserver-driven reveal animations and active nav link logic

Accessibility & Semantics

6/10

Usable for everyone

Solid a11y basics and semantics - skip link, ARIA on dialog and live regions, keyboard handlers and focus restoration - applied to custom UI rather than inherited from a kit.

Evidence

portfolio/index.html: skip-link, dialog element with aria-labelledby and focus restore on close

portfolio/index.html: toast with aria-live='polite', accessible menu button with aria-expanded and keyboard Escape handling

State Management & Data Flow

3/10

Managing data in the app

Server-state handling is pragmatic: a single custom hook with demo fallback, error logging and loading state but no advanced cancellation, optimistic-update rollback or state machine for complex flows.

Evidence

Interview_Prep-Tracker/src/hooks/usePrepData.ts: supabase-backed CRUD, demo-mode fallback and loading/error handling

Interview_Prep-Tracker/src/lib/supabase.ts: client initialization with environment checks

UX & Visual Polish

5/10

Look and feel quality

High visual polish and interaction design with deliberate transitions, reveal animations and micro-interactions; UI lacks widespread explicit loading skeletons and undo/rollback UX for server ops.

Evidence

portfolio/index.html: count-up, marquee, reveal animations and animated project dialogs

Interview_Prep-Tracker/src/index.css: theme tokens and Excalidraw container styling supporting embedded diagrams

Expertise

React· Middle

Frontend Architecture & Build Tools· Middle

Web Performance & Optimization· Middle

Industries

Artificial Intelligence· Middle

Software· Middle

Financial Services· Middle

Technologies

JavaScript

TypeScript· since 2024 · Middle

Tailwind CSS

Next.js

React.js

Vite

PostCSS

Recharts

autoprefixer

Recommendations

Lead development of customer-facing, interactive web apps that require accessible modals, keyboard navigation and smooth animations.
Build a component library and Storybook from existing UI patterns to increase reuse and make accessibility guarantees explicit.
Own frontend build-tooling and performance audits (bundle analysis, code-splitting, LCP/INP improvements) for single-page apps.
Expand server-state patterns by adding request cancellation, optimistic updates with rollback and unit/integration tests around critical flows (usePrepData hook as a starting point).

Repositories

The developer's experience in this domain has been verified based on AI analysis of the following repositories:

portfolio Interview_Prep-Tracker VivekanandChoudhary09

Job preferences

Work permits: India
Citizenship: India
Salary: from $40k per year
Employment type: Full-Time