Mohd Zeeshan profile photo 1 on Tal

Mohd Zeeshan

Data Scientist | MS by Research at IIIT-Hyderabad | Computer Vision & GenAI | ex-Akridata

Hyderabad

2+ years of work experience

Akridata
Oct 20241 yr 7 mosApr 2026

Applied Vision-Language Research for Industrial Visual Inspection Deployed a fine-tuned Multimodal SLM in production for a medical device manufacturing client which increased accuracy from 60% to 98% and inference speed from 0.1 to 4 FPS with just 20 training images Created and deployed an end-to-end LoRA fine-tuning and inference API server for internal agentic experiments. Deployed GAN-based anomaly detection model in production Researched on Diffusion models for synthetic defect generation.
Akridata
May 20245 mosSep 2024

Tested and LoRA fine-tuned zero-shot object detection VLMs (Vision Language Models) Added a new model to production, increasing zero-shot object detection accuracy by ~5%
iHub-Data, IIIT Hyderabad
Mar 20243 mosMay 2024

Hyderabad, Telangana, India

iHub-Data, IIIT Hyderabad
Dec 20234 mosMar 2024

On-site

AI/ML program @ IIITH
T-Hub
Jul 20236 mosDec 2023

Telangana, India

Part of T-hub's RubriX 3.0 program

Mohd’s AI Stack

hugging face
hugging face
langchain
langchain
ollama
ollama
openai
openai

Education

International Institute of Information Technology Hyderabad (IIITH)

Master of Science - MS, Computer Science

International Institute of Information Technology Hyderabad (IIITH)

20252 yrs2027
Osmania University

Bachelor of Engineering - BE, Computer Science

Osmania University

20204 yrs2024

Join Mohd and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Mohd Zeeshan on Tal

Headline

Data Scientist | MS by Research at IIIT-Hyderabad | Computer Vision & GenAI | ex-Akridata

About

As a Data Scientist at Akridata, I contributed to cutting-edge industrial visual inspection solutions by deploying advanced Vision-Language Models (VLMs) and Multimodal LLMs. Successfully implemented a fine-tuned multimodal SLM that improved defect detection accuracy from 60% to 98% for a medical device manufacturing client. Delivered impactful results by enhancing inference speed and optimizing performance with limited training data. Additionally, developed and deployed end-to-end LoRA fine-tuning and inference API servers, enabling agentic experiments. Currently pursuing a part-time MS in Computer Science at IIIT Hyderabad, with a research focus on computer vision and machine learning. Holds a Bachelor of Engineering in Computer Science from Osmania University.

Location

Hyderabad

AI stack

hugging face

langchain

ollama

openai

Work experience

Data Scientist-1 at Akridata, Oct 2024 - Apr 2026 · 1 yr 7 mos, Applied Vision-Language Research for Industrial Visual Inspection Deployed a fine-tuned Multimodal SLM in production for a medical device manufacturing client which increased accuracy from 60% to 98% and inference speed from 0.1 to 4 FPS with just 20 training images Created and deployed an end-to-end LoRA fine-tuning and inference API server for internal agentic experiments. Deployed GAN-based anomaly detection model in production Researched on Diffusion models for synthetic defect generation.

Machine Learning Intern at Akridata, May 2024 - Sep 2024 · 5 mos, Tested and LoRA fine-tuned zero-shot object detection VLMs (Vision Language Models) Added a new model to production, increasing zero-shot object detection accuracy by ~5%

SRISHTI-24 Research Intern at iHub-Data, IIIT Hyderabad, Mar 2024 - May 2024 · 3 mos, Hyderabad, Telangana, India

Student Trainee at iHub-Data, IIIT Hyderabad, Dec 2023 - Mar 2024 · 4 mos, On-site, AI/ML program @ IIITH

Member at T-Hub, Jul 2023 - Dec 2023 · 6 mos, Telangana, India, Part of T-hub's RubriX 3.0 program

Education

Master of Science - MS, Computer Science, International Institute of Information Technology Hyderabad (IIITH)

Bachelor of Engineering - BE, Computer Science, Osmania University

Mohd Zeeshan - Data Scientist-1 in Hyderabad | Tal