Available · 3 Months Notice · London, UK

Vinay KumarK V

Applied AI/ML Practitioner · GenAI & Multimodal AI · GxP Regulated AI · Doctoral Researcher in VLA Models & Embodied AI

I build AI systems in regulated environments — and I've been doing it at GSK. Contributed to LLM evaluation for regulatory document automation, supported RAG pipeline POC design, and deployed production Python tools in live pharmaceutical supplier workflows. Independently building VATSA — a unified five-modality AI architecture — with a published preprint on Zenodo (April 2026). Doctoral research focus: VLA models for safety-critical autonomous systems. Long-term mission: a safe embodied AI that walks amongst humans.

8+
Years in IT & pharmaceutical regulated environments
96%
CIFAR-10 accuracy — VATSA Visual Encoder (EfficientNet-B0)
1
Published preprint — VATSA architecture (Zenodo, Apr 2026)
22
FPS real-time object detection — YOLOv8 integration
// 01 — About

Where domain expertise
meets AI engineering

I am an Applied AI/ML Practitioner with 8+ years of experience across IT and GxP-regulated pharmaceutical environments, actively transitioning into hands-on AI/ML engineering. My background spans electronics engineering, enterprise systems, an MBA, and now a doctoral programme in AI & ML.

At GSK, I contributed to AI product development for regulatory document automation — defining evaluation criteria for LLM comparative analysis (Azure Document Intelligence vs GPT-4o), collaborating on RAG pipeline POC design, and deploying production Python utilities used in live regulated supplier workflows.

That experience made one thing clear: I did not want to keep specifying AI systems. I wanted to build them. So I invested deliberately — a DBA in AI & ML delivered via Great Learning (Walsh College, 2025–2028), with Year 1 at Texas McCombs School of Business, alongside hands-on independent project building.

My long-term research focus is AI in Robotics — specifically Vision-Language-Action (VLA) models for autonomous systems in safety-critical environments. Independently building VATSA — a unified five-modality AI architecture (Video, Audio, Text, Sensory, Action) — with a published preprint on Zenodo (April 2026) and a proposed novel output routing mechanism, SAMOS (Safety-Aware Multi-Output Selector).

The long-term mission: a safe embodied AI that walks amongst humans — capable but fundamentally correctable, transparent, and subordinate to human intent.

"What I bring that most AI engineers cannot: eight years of understanding how regulated businesses make decisions, how to frame the right problem before touching a line of code, and how to communicate technical trade-offs to a board-level audience — combined with genuine hands-on AI development."

ROLEApplied AI/ML PractitionerTransitioning to AI/ML Engineer
BASEDLondon, United KingdomUK Work Permit · Open to relocation
STUDYDBA in AI & ML — Walsh Collegevia Great Learning · Year 1: Texas McCombs PGP
RESEARCHVLA Models · Embodied AI · VATSASAMOS — Safety-Aware Multi-Output Selector
DOMAINPharma · Regulated AI · GxP6+ years GSK pharmaceutical systems
OPEN TOAI/ML Engineer · Applied AI · GenAIAny industry solving real problems with AI
// 02 — Skills

Technical Stack

GenAI & Agentic AI
LLMs & Prompt Engineering RAG Architecture LangChain Multi-Agent Systems FAISS · Vector Databases Azure OpenAI Gemini API Embeddings LangGraph (learning) Fine-tuning (learning) HuggingFace Transformers (learning)
Deep Learning & Computer Vision
PyTorch TensorFlow EfficientNet · Transfer Learning CNNs · RNNs · LSTM YOLOv8 · Object Detection Embeddings · Latent Space Real-time Video Streams Audio Encoding — Wav2Vec · Whisper (learning) Multimodal Fusion (learning)
Core DS & ML
Python Pandas · NumPy · Scikit-learn NLP & Text Classification XGBoost · Random Forest · Ensemble Methods EDA & Feature Engineering Explainable AI (SHAP) Statistical Analysis SQL Streamlit · FastAPI · Gradio MLflow (learning)
Cloud & Deployment
Azure OpenAI · Document Intelligence Azure Container Apps · ACR Docker Git · GitHub · CI/CD HuggingFace Spaces
Regulated AI & Research
GxP · Regulated AI Responsible AI · AI Governance URS · SRS · UAT Documentation Workshop Facilitation Stakeholder Management Agile · Scrum VATSA Architecture · SAMOS VLA Models (DBA research) Reinforcement Learning (learning)
// 03 — Experience

Where I've contributed

May 2023 – Present
TCS · GSK plc
London, UK
Technical Business Analyst — AI/ML Focus
  • Defined domain requirements and extraction quality criteria for LLM comparative evaluation (Azure Document Intelligence vs GPT-4o) — contributed to recommendation that shaped regulatory document automation strategy across US and UK
  • Collaborated on RAG pipeline POC design alongside the technical team — document retrieval strategies, prompt templates, and factuality requirements in a GxP-regulated environment
  • Built and deployed Python utilities in live production at GSK — XML digital signature verification tools and Excel-to-XML converters used in regulated pharmaceutical supplier workflows
  • Automated Jira sprint analytics pipeline using Python and Jira API — eliminated manual reporting effort for a 25+ member cross-functional team
  • Facilitated business workshops across QA, Operations, IT, and Regulatory teams — translated AI evaluation findings into business recommendations securing executive buy-in
  • Authored full GxP documentation suite (URS, SRS, UAT protocols, validation reports) ensuring regulatory compliance and reproducibility across AI tooling
  • Received multiple internal recognitions for delivery and stakeholder impact
Sep 2019 – May 2023
TCS · GSK plc
Bangalore & London
Business Analyst & Scrum Master
  • Progressed from Validation Lead to BA and Scrum Master — led overall product development with a cross-functional team of 8+ developers across 8 global pharmaceutical manufacturing sites
  • Converted business requests into structured problem statements with measurable acceptance criteria — reduced rework across 8+ major releases
  • Owned GxP validation documentation end-to-end throughout the product development lifecycle
  • Led full transition from development to service support with structured knowledge transfer — received formal customer appreciation
Oct 2015 – Oct 2017
TCS · Telefonica
Chennai, India
Systems Engineer
  • Maintained enterprise backup and restore infrastructure across Linux and Windows environments — 100% SLA compliance
  • Built shell scripting automation tools reducing manual operations and improving incident response time
// 04 — Projects

Things I've built

PROJECT 01
AI Doc2XML — Dual-Agent System Containerised

Independently built agentic AI system using LangChain orchestrating two specialised agents — an Extractor and a Reviewer — for pharmaceutical regulatory document processing. Containerised with Docker and deployed live on Azure Container Apps.

PythonLangChainMulti-AgentDockerAzure Container AppsGradio
PROJECT 02
GxP Quality Intelligence — NLP + XAI

End-to-end NLP classifier for pharmaceutical deviation categorisation (critical/major/minor). TF-IDF feature engineering, Logistic Regression, FAISS cosine retrieval, and Explainable AI layer using SHAP for GxP regulatory audit acceptance.

PythonScikit-learnTF-IDFFAISSSHAP / XAINLP
PROJECT 03
Customer Churn Prediction Live on HuggingFace

End-to-end ML project addressing class imbalance in churn prediction. Deployed live on HuggingFace Spaces with Streamlit UI and FastAPI backend. Evaluated using F1 and AUC-ROC on minority class — not accuracy.

PythonScikit-learnStreamlitFastAPIHuggingFaceSMOTE
PROJECT 04
Jira Sprint Reporter — Python Automation

Python and Jira API automation tool with sprint analytics, automated email delivery, and visual reporting. Deployed live at GSK — eliminated manual reporting effort for a 25+ member cross-functional team.

PythonJira APIPandasAutomationGit
RESEARCH PROJECT — IN PROGRESS
VATSA — Video · Audio · Text · Sensory · Action V-Module Complete

Unified five-modality AI architecture for human-level perception and action. Each modality encoder projects into a shared 512-dim latent space for cross-modal fusion. V-Module complete: EfficientNet-B0 fine-tuned to 96.31% accuracy on CIFAR-10, integrated with YOLOv8 for real-time object detection at 22 FPS. Benchmarked at 1,336 embeddings/sec at batch 16. Proposes SAMOS (Safety-Aware Multi-Output Selector) — a novel output routing mechanism using asymmetric safety-weighted sigmoid thresholds per modality, enabling safe parallel multi-modal output generation in physically embodied AI systems. Architecture published as a preprint on Zenodo (April 2026).

PyTorchEfficientNet-B0Transfer LearningYOLOv8Multimodal AIEmbeddingsSAMOSVLA Models
View All Projects & Deep Dives →
// 04b — Publications

Research Output

Preprint · April 2026 · Zenodo
VATSA: Video, Audio, Text, Sensory, Action — A Unified Five-Modality Architecture for Human-Level Perception and Action
Vinay Kumar K V · DBA in AI & ML (Walsh College / Texas McCombs via Great Learning)

Presents the VATSA conceptual architecture and four core principles: shared latent space, cross-modal attention, temporal coherence, and closed-loop action. Introduces SAMOS (Safety-Aware Multi-Output Selector) — a novel output routing mechanism using asymmetric safety-weighted sigmoid thresholds for safe parallel multi-modal generation in embodied AI systems. Covers motivating applications in healthcare, pharma, autonomous systems, and adaptive education.

View on Zenodo → Download PDF →
// 05 — Education

Academic foundation

Master of Business Administration
MSRIT, Bangalore
2017 – 2019
Marketing & HR. Business problem framing, ROI analysis, stakeholder communication. Project: Market Research.
BE — Electronics & Communication Engineering
Sir M Visvesvaraya Institute of Technology
2011 – 2015
Foundation in control systems, sensors, and signal processing — directly relevant to robotics research direction. Project: Smart Railways — Automated Railways.
// 06 — Contact

Let's connect

Open to AI/ML Engineer, Applied AI, and GenAI roles across any industry where AI is the core solution. Available with 3 months notice.

Email
vinaykumar.kv
@outlook.com
LinkedIn
vinay-kumar-k-v
GitHub
vinaykumarkv