Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Hi, I’m Velivela Rohit Kumar, a junior AI engineer with 1 year of hands-on experience delivering production-grade AI solutions in cloud environments. I’m proficient with LLMs, transformers, Retrieval-Augmented Generation (RAG), and advanced prompt engineering to deliver high-quality outputs, and I’ve built Flux.dev image pipelines with custom LoRAs and IP adapters for consistent visuals. I enjoy creating end-to-end ML pipelines and cloud-native deployments on AWS, with a focus on scalable architectures, robust testing, and clean data contracts. Notable work includes Pitch Craft A and Cine Sketch AI—tools that turn scripts into pitch decks and storyboards—where I integrated LLM-generated text, image generation, and PDF exports while ensuring character consistency and reproducibility.…Hi, I’m Velivela Rohit Kumar, a junior AI engineer with 1 year of hands-on experience delivering production-grade AI solutions in cloud environments. I’m proficient with LLMs, transformers, Retrieval-Augmented Generation (RAG), and advanced prompt engineering to deliver high-quality outputs, and I’ve built Flux.dev image pipelines with custom LoRAs and IP adapters for consistent visuals. I enjoy creating end-to-end ML pipelines and cloud-native deployments on AWS, with a focus on scalable architectures, robust testing, and clean data contracts. Notable work includes Pitch Craft A and Cine Sketch AI—tools that turn scripts into pitch decks and storyboards—where I integrated LLM-generated text, image generation, and PDF exports while ensuring character consistency and reproducibility.

Velivela Rohit Kumar

Data Scientist, AI Developer, Full Stack Developer, +3





Hi, I’m Velivela Rohit Kumar, a junior AI engineer with 1 year of hands-on experience delivering production-grade AI solutions in cloud environments. I’m proficient with LLMs, transformers, Retrieval-Augmented Generation (RAG), and advanced prompt engineering to deliver high-quality outputs, and I’ve built Flux.dev image pipelines with custom LoRAs and IP adapters for consistent visuals. I enjoy creating end-to-end ML pipelines and cloud-native deployments on AWS, with a focus on scalable architectures, robust testing, and clean data contracts. Notable work includes Pitch Craft A and Cine Sketch AI—tools that turn scripts into pitch decks and storyboards—where I integrated LLM-generated text, image generation, and PDF exports while ensuring character consistency and reproducibility.…Hi, I’m Velivela Rohit Kumar, a junior AI engineer with 1 year of hands-on experience delivering production-grade AI solutions in cloud environments. I’m proficient with LLMs, transformers, Retrieval-Augmented Generation (RAG), and advanced prompt engineering to deliver high-quality outputs, and I’ve built Flux.dev image pipelines with custom LoRAs and IP adapters for consistent visuals. I enjoy creating end-to-end ML pipelines and cloud-native deployments on AWS, with a focus on scalable architectures, robust testing, and clean data contracts. Notable work includes Pitch Craft A and Cine Sketch AI—tools that turn scripts into pitch decks and storyboards—where I integrated LLM-generated text, image generation, and PDF exports while ensuring character consistency and reproducibility.

Available to hire

Hi, I’m Velivela Rohit Kumar, a junior AI engineer with 1 year of hands-on experience delivering production-grade AI solutions in cloud environments. I’m proficient with LLMs, transformers, Retrieval-Augmented Generation (RAG), and advanced prompt engineering to deliver high-quality outputs, and I’ve built Flux.dev image pipelines with custom LoRAs and IP adapters for consistent visuals.

I enjoy creating end-to-end ML pipelines and cloud-native deployments on AWS, with a focus on scalable architectures, robust testing, and clean data contracts. Notable work includes Pitch Craft A and Cine Sketch AI—tools that turn scripts into pitch decks and storyboards—where I integrated LLM-generated text, image generation, and PDF exports while ensuring character consistency and reproducibility.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

English

Advanced

Work Experience

Junior AI/ML Developer at Quantum AI Global

August 1, 2024 - Present

Built Pitch Craft A, a tool that converts film scripts into pitch decks and visual storyboards, using LLM-generated text and Flux image generation with character consistency; designed microservices in Python/FastAPI with versioned REST routes, JWT auth, and PostgreSQL persistence for projects/assets; orchestrated AI pipelines via a secondary service: Fireworks API for text and Flux endpoints for images, with unified JSON contracts and robust error handling.

Jr AI/ML Developer at Quantum AI Global

August 1, 2024 - Present

Built Pitch Craft A tool converting film scripts into pitch decks and visual storyboards using LLM-generated text and Flux image generation with character consistency; designed microservices in Python/FastAPI with versioned REST routes, Pydantic schemas, JWT auth, and PostgreSQL persistence for projects/assets; orchestrated an AI pipeline via a secondary service (Fireworks API for text and Flux endpoints for images) with unified JSON contracts, robust error handling, and structured logging; drove SDLC best practices end-to-end including requirements from BRS, sprint planning, code reviews, CI/CD, unit/integration testing, release management, and documentation; applied advanced prompt engineering with custom LoRAs and IP-Adapters to ensure consistent, reproducible visuals; authored BRS-aligned technical design and data contracts; mapped the flow (project → shots/prompts → images → editor → PDF) for scope control and traceability; delivered Cine Sketch AI, a tool turning scripts

Intern, Machine Learning Engineer at Quantum AI Global

January 1, 2024 - July 1, 2024

Delivered HealMed, a GenAI mental-health assessment tool for organizations; analyzes behavioral, sentiment, and contextual signals to produce counselor-ready insights and recommendations. Designed role-based onboarding and dashboards (Org, HR, Counselor, Employee) with a questionnaire builder aligned to PHQ/GAD, real-time previews, and region/domain customization via a central question database. Implemented multimodal AI: real-time audio/video emotion and sentiment analysis, speech-to-text, and adaptive follow-up generation with llama-3, including randomization to reduce order bias. Built LangChain + llama-3 agents to synthesize responses and A/V signals into prescriptive, visual reports; added a reports portal for counselor review, download, and categorization. Established security & governance: unified counselor data platform with secure access controls and encryption; anonymized storage of A/V inputs to meet privacy requirements.