Engineering AI Systems
With Humans in Mind

I'm a GenAI and Agentic AI Engineer with a strong applied research orientation, building production-grade intelligent systems where human control, safety, and trust aren't afterthoughts—they're the foundation.

Vishwanath Rajasekaran

My Approach

My work spans LLM-powered applications, retrieval-augmented and multi-agent workflows, and end-to-end platforms—with hands-on ownership from system architecture and model orchestration through deployment, evaluation, and continuous iteration in real environments.

Across projects in education, mental-health decision support, social platforms, and knowledge systems, I conduct applied research on interaction design, explanation structure, autonomy boundaries, and user behaviour, using empirical analysis to inform system design.

I prioritise cost-aware AI engineering, implementing model routing, caching, and tiered inference strategies to balance quality, latency, and spend—delivering systems that are scalable, auditable, economically sustainable, and built to work effectively with humans, not around them.

Safety First

Building AI systems that fail gracefully and keep humans in control

Human-Centered

Designing interactions that calibrate trust and preserve autonomy

Production-Ready

End-to-end systems that scale, audit, and operate reliably

Who I Am

I believe technology should elevate humanity, not replace what makes us human.

Outside of engineering AI systems, I'm a committed fitness enthusiast who never skips the gym—discipline in one area fuels discipline everywhere. I'm deeply curious about different cultures, love meaningful conversations, and find myself naturally drawn to understanding what drives people emotionally.

Family comes first, always. That grounding shapes how I think about building technology: it should serve people, help them grow, and make space for what truly matters—pursuing passions, connecting with loved ones, and spreading a little more kindness in the world.

I carry a quiet conviction that one day, AI will handle most of our work. But even then, we'll need purpose beyond productivity. My goal is to build systems that free people to explore, create, and live fully—not systems that make us dependent or idle.

💪

Discipline & Consistency

Never skip the gym, never skip the grind

🌍

Cultural Curiosity

Learning from every corner of the world

👨‍👩‍👧

Family First

The foundation that grounds everything

🤝

Human Connection

Understanding emotions, building bridges

🌱

Tech for Good

Making Earth a better place through innovation

Purpose-Driven

Passion over productivity, always

Technical Skills

🤖 GenAI & Agentic AI

  • Production LLM Systems (OpenAI, Claude, Llama)
  • Agentic Workflows & Multi-Agent Coordination
  • Retrieval-Augmented Generation (RAG & GraphRAG)
  • Context, Memory & State Management
  • MCP-Style Orchestration
  • Tool Invocation & Function Calling

🔬 Applied AI Research

  • Human–AI Interaction Analysis
  • Trust Calibration & Autonomy Boundaries
  • Explainability (XAI) & Transparency
  • Interaction-Log Analysis
  • Failure-Mode Analysis (Hallucination, Over-trust)
  • Safety-Aware & HITL System Design

💰 Cost-Aware AI Engineering

  • Model Routing & Tiered Inference
  • Caching & Fallback Strategies
  • Throughput Optimisation
  • LLM Cost Monitoring
  • Budget-Aware Deployment
  • Scaling Under Business Constraints

Full-Stack Development

  • Backend: Flask, FastAPI, Django
  • Frontend: React, Next.js
  • Languages: Python, JavaScript, TypeScript
  • APIs: REST, GraphQL
  • AI-Integrated Web Platforms
  • End-to-End Ownership

🗄️ Knowledge & Data Systems

  • Vector DBs: FAISS, Pinecone, ChromaDB
  • Graph DBs: Neo4j, Cosmos DB
  • Relational: PostgreSQL, MySQL
  • Document Databases (MongoDB)
  • Schema-Aware Retrieval
  • Structured Reasoning

☁️ Cloud & DevOps

  • AWS: EKS, EC2, Lambda, S3, API Gateway
  • Azure: App Services, Cosmos DB, Blob
  • Docker & Kubernetes (EKS/AKS)
  • Terraform, Jenkins, GitHub Actions
  • Cloud Monitoring: CloudWatch, Azure Monitor
  • OSS Observability: Prometheus, Grafana
  • IAM & RBAC, Secrets Management & Hardening

📊 ML & Evaluation

  • Classical ML (Gradient Boosting, Supervised)
  • Experimentation & Tracking (MLflow)
  • Output Quality & Drift Monitoring
  • Prompt & Retrieval Evaluation
  • Robustness Testing

🛡️ Trust, Safety & Governance

  • Responsible AI Design
  • Privacy-First Architecture
  • Role-Based Access Control
  • Moderation Workflows
  • Auditability & Logging
  • Oversight Mechanisms

🧭 Engineering Leadership

  • Technical Decision Ownership & Trade-offs
  • Architecture & Delivery Leadership
  • Mentorship & Technical Guidance
  • Roadmap Planning & Prioritisation
  • Stakeholder & Cross-Team Collaboration
  • Risk, Cost & Quality Accountability

Engineering & Research Journey

2024 — Present

Lead Full-Stack AI/ML Engineer

Smartail UK

Leading the design and delivery of production GenAI systems, including LLM-powered applications, RAG and GraphRAG pipelines, and agentic workflows. Owning architecture decisions, cost and safety trade-offs, cloud deployment, and cross-team delivery.

2023 — Present

Freelance & Independent AI Projects

Designing and delivering applied AI and GenAI systems alongside full-time work, including conversational assistants, retrieval-augmented pipelines, and agentic workflows, with emphasis on reliability, safety, and real-world usability.

2023 — Present

Human–AI Interaction Research Direction

Industry-Aligned Research Focus

Developing a research focus on human–AI interaction, trust calibration, autonomy boundaries, and explanation design, informed by MSc research and ongoing industry systems work.

2023 — 2024 · MSc Computer Science (Merit)

MSc Dissertation — Mental-Health AI Prediction

Coventry University, United Kingdom

Designed and implemented a full-stack mental-health prediction system integrating machine learning with human-centred interface design. Studied how users interpret, trust, and act on AI predictions in high-sensitivity contexts.

2019 — 2022

Senior Software Engineer

Accenture

Built and deployed production backend and ML systems using Python and cloud-native architectures. Led API development, CI/CD automation, and cloud migration initiatives while collaborating closely with cross-functional teams.

2015 — 2019

BTech — Information Technology

Sri Krishna College of Engineering and Technology, India

Built a strong foundation in computer science and software engineering, covering data structures, databases, operating systems, and web technologies. Developed early interest in human–computer interaction through applied projects, programming, and system design.

Interested in Working Together?

I'm open to AI engineering roles, research positions, and collaborative projects in human-AI interaction.