Ankit Gupta

Lead ML Engineer

Delhi, India5 yrs 10 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Expert in fine-tuning large language models.
Proficient in developing scalable AI solutions.
Strong background in machine learning and deep learning frameworks.

Stackforce AI infers this person is a skilled AI Research Engineer specializing in Generative AI and Machine Learning.

Contact

Skills

Core Skills

Generative AiMachine LearningLarge Language Models (llm)Python (programming Language)

Other Skills

Multi Modal ModelsSpeculative DecodingPyTorchC++Retrieval-Augmented Generation (RAG)Database Management System (DBMS)Data StructuresMicrosoft OfficeMicrosoft ExcelManagementMicrosoft WordBlenderMicrocontrollersIntel 8085

About

I am a dedicated software professional specializing in applied AI research, with expertise in C/C++ and Python, and proficiency in frameworks for machine learning and deep learning. My experience includes fine-tuning large language models (LLMs), developing Retrieval-Augmented Generation (RAG) systems, and implementing classical ML and DL algorithms. Through hands-on projects, I have honed my skills in creating scalable AI solutions, emphasizing innovation and problem-solving. My commitment to continuous learning enables me to tackle complex challenges and contribute effectively to cutting-edge, technology-driven AI research environments.

Experience

5 yrs 10 mos

Total Experience

2 yrs 11 mos

Average Tenure

4 yrs

Current Experience

Qualcomm

3 roles

Senior Machine Learning Engineer

Promoted

Nov 2024 – Present · 1 yr 6 mos · Hyderabad, Telangana, India

Implemented Model Sharing and Switching across process, to eliminate resource duplication and memory optimization.
Improved Prompting/Decoding by 1.2x for SLM architecture models. Reduced memory footprint by 0.2x.
Implemented ACTIONS like ABORT, PAUSE, RESUME for GenAI models in GENIE.
Worked on Speculative Decoding Methods like SSD, LADE, SPD to improve decoding rate by 1.5-2.2x over standard AR decoding.
Worked towards catering RAG use case for BGE Large models.
Worked on multi-modal models for image captioning and story generation.
Worked on adding LoRA capabilities to GENIE.
Working on optimising and improving more GeNAI usecases for Qualcomm NPU.

Generative AIMachine Learning