Ashly Ajith

Ashly Ajith

Senior Data Scientist

Mumbai, India · Mahindra AI Division

Senior AI/ML professional with 4.5+ years building production GenAI systems at Mahindra. Specialising in Agentic AI, LLM orchestration, and real-time AI products that serve millions.

GenAIAgentic AILLM OrchestrationProduction MLVoice AIRAG

The Full Story

I'm Ashly Ajith — a Senior Data Scientist and AI/ML leader currently driving AI innovation at the Mahindra AI Division in Mumbai.

With 4.5+ years of hands-on experience across enterprise AI product development, I build systems that go beyond proof-of-concept into real production — serving real customers at scale. My work spans the full AI stack: from fine-tuning LLMs and deploying SLMs on-premise, to orchestrating multi-agent systems and building real-time voice AI pipelines.

At Mahindra, I've led the development of Naina — a multilingual real-time AI voice assistant serving 1 lakh+ automotive customers — and architected a Multi-Agent AI platform that automates hyper-personalized marketing creative generation across the Mahindra brand portfolio. Prior to this, I built the PDIT Engineering Copilot, an org-wide RAG-powered AI assistant embedded into Microsoft Teams for Mahindra's engineering teams.

I hold an M.Tech in Artificial Intelligence from Amrita School of Engineering (GPA 9.63, Rank Holder) and am AWS Cloud Practitioner certified. My research in computer vision and domain adaptation has been published in international venues.

I'm passionate about the intersection of applied AI research and enterprise product engineering — building systems that are robust, explainable, and genuinely useful.

Professional Experience

Senior Data Scientist – Manager

Mahindra AI Division

Mumbai, India

Sept 2024 – PresentCurrent
  • Led development of Naina — multilingual real-time voice AI assistant serving 1L+ customers
  • Architected Multi-Agent AI platform for hyper-personalized marketing creative generation
  • Technical lead for personalized user journey model targeting 10% sales uplift in Mahindra Automotive
  • Led prototyping of generative media content model using Vision Transformers
  • Manages cross-functional teams, vendor negotiations, client demos, and hackathons

Deputy Manager – AI/ML Digital Lead

Mahindra and Mahindra (Automotive & Farm Division)

Chennai, India

July 2022 – Aug 2024
  • Built and deployed PDIT Engineering Copilot — org-wide AI assistant in Microsoft Teams
  • Automated vehicle HMI testing using computer vision and OCR pipelines
  • Developed PLM Syslog Analytics using DBSCAN anomaly detection and Power BI dashboards
  • Designed custom Named Entity Recognition (NER) for automotive ORC product reports

Research Trainee

Nokia Solutions and Networks

Bangalore, India

July 2021 – Aug 2022
  • Built NLP-powered RCA classifier with 83% accuracy for real-time ticket analysis
  • Reduced 2-hour RPA pipeline to 8 seconds using Selenium-based automation
  • Curated and cleaned 500+ data point RCA dataset using scraping tools

Technical Skills

genai

LLMsRAGAgentic AILLM Evaluation (RAGAS)Prompt Engineering

agents

CrewAIAutogenLangGraphOpenAI Functions

finetuning

UnslothGemma 3Llama 3.1Phi-4Voice CloningPyTorch

mlops

Azure ML StudioAWS BedrockAWS SageMakerGCP Vertex AIModel Monitoring

nlp

BERTCustom NERWord2VecTF-IDF

classical

RegressionClassificationClusteringTime SeriesDeep Learning

deployment

FastAPIFlaskStreamlitAzure Bot ServicesDocker

Education

M.Tech — Artificial Intelligence

Amrita School of Engineering

2020 – 2022

GPA 9.63Rank Holder, S2 Merit Scholarship

B.Tech — Electrical and Electronics Engineering

School of Engineering, CUSAT

2015 – 2019

Certifications

Deep Learning and Reinforcement Learning Specialisation

DeepLearning.AI / IBM

AWS Cloud Practitioner

Amazon Web Services

Certified Scrum Master (CSM) & A-CSM

Scrum Alliance

Competent Communicator

Toastmasters International