Piyush Aaryan

Founder

Delhi, India2 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Expert in fine-tuning LLMs for Indic languages.
Saved clients over $100k on cloud deployment costs.
Developed innovative multimodal AI solutions.

Stackforce AI infers this person is a skilled AI Engineer with a focus on cost-effective solutions in the AI industry.

Contact

Skills

Core Skills

Generative AiArtificial Intelligence (ai)

Other Skills

Large Language Models (LLM)cloud servicescost managementopen-source LLMsdatabase queryingmultimodal RAGdata analysisSoftware AgentsGo (Programming Language)GolangStable Diffusionvector databaseComputer VisionImage SegmentationStatistical Concepts

About

Currently pursuing a BTech in CS-AI at Maharaja Agrasen Institute of Technology, while contributing as an AI & Backend Developer on Upwork. Successfully collaborated on projects involving fine-tuning LLMs for Indic languages and reducing cloud deployment costs by leveraging cost-effective GPUs over AWS SageMaker. Developed innovative solutions such as database querying using open-source LLMs and multimodal retrieval-augmented generation (RAG) for analyzing diverse data formats. Passionate about advancing AI applications in cost-efficient ways, with expertise in Golang, Python, and software agents.

Experience

Versely

Founder

Jan 2026 – Present · 2 mos · Gurugram · On-site

32nd

AI Intern

May 2024 – Jul 2024 · 2 mos · Gurugram, Haryana, India · Hybrid

Artificial Intelligence (AI)Software Agents

Upwork

AI Engineer (Top Rated)

Jan 2024 – Present · 2 yrs 2 mos · United States · Remote

An AI Engineer earned $45k+, previously worked with different clients and finetuned llms on indic languages. I have saved $100k+ for my clients on cloud services while deploying the model on cost effective gpus instead aws sagemaker. I have used rag to build different projects like to query from databases using open-source llm like zephyr 7b beta to save cost. Multimodal rag to query from the images/tables/graphs from pdfs, docs etc using the opensource multimodal llm instead of gpt4v to save cost. Virtual try-on , mannequins to models , teeth alignment with Gans etc.

Generative AIArtificial Intelligence (AI)