Explicit boundaries
Define interfaces, ownership and data flow.
SOFTWARE ENGINEERING × AI / ML PORTFOLIO / 2026
Mridul Singh Saklani
Software and AI/ML Engineer
Software and AI/ML Engineer.
Backend systems. Applied AI. Cloud architecture.
Grounded in physics and mathematics.
01ENGINEERING APPROACH
01SYSTEMS THINKING
Software systems. Applied intelligence.
Software and AI/ML Engineer with a physics background. I build backend services, data workflows and AI applications.
INTERFACES. DATA FLOW.
FAILURE HANDLING.
FAILURE HANDLING.
PROFILE / 001SOFTWARE + AI/ML
I design around explicit boundaries, data flow and failure handling. Reliability and observability are part of the architecture.
My technical focus spans backend services, cloud architecture and applied AI: retrieval, language models, agents, computer vision and automation.
My interests extend to physics, quantum mechanics, aerospace, mathematics, history and geopolitics.
AI in context
Connect models to retrieval, tools and application constraints.
Failure-aware design
Plan for retries, visibility and recoverable faults.
01 / PHYSICS
Bachelor’s in Physics
Himachal Pradesh University
Scientific reasoning, mathematics and physical systems.
02 / COMPUTER APPLICATIONS
Master of Computer Applications
Lovely Professional University
Software development, algorithms and computing systems.
01 / 07DESIGN · IMPLEMENT · MEASURE · REFINE
02APPLIED AI
Models, retrieval
and tool execution.
Models operate within systems. Context, tools, APIs and evaluation determine how those systems behave.
SELECT A TOPICINTERACTIVE / 05 NODES
NODE / 01
LLMs & GenAI
Language models integrated with application logic, data and APIs.
GENERATIVE AILLMSRAGMCPCOMPUTER VISIONAI AGENTS
03CLOUD / SYSTEM DESIGN
Services.
Data. Operations.
Service boundaries, asynchronous workloads and the operational constraints that shape an architecture.
AWS SERVICE GROUPS · INTERACTIVESERVICE MAP / 01SELECT A SERVICE GROUP INTERACTIVE MAP
NODE / 01
Edge & delivery
Object storage and edge delivery for application content. Services: CloudFront, S3.
An overview of service families. Project-specific architectures are documented with their implementations.
04ENGINEERING DISCIPLINES
Core
engineering disciplines.
Software, data, infrastructure and AI. Select a discipline for its tools and technical scope.
TECHNICAL
SCOPE
SCOPE
Interfaces between
engineering disciplines.
NODE / 01
AI / ML systems
Language models, retrieval, agents and computer vision. Areas: Generative AI · LLMs · RAG · MCPs · AI agents · Computer vision · Intelligent automation.
FASTAPI · PYTHON · NODE.JS · REACT / NEXT.JS · REST APIS · POSTGRESQL · MONGODB · ELEVENLABS
05TECH STACK
The technical
stack.
Technologies grouped by their role in an application. Detailed notes are available on the skills page.
Languages
03- Python
- JavaScript
- TypeScript
Frontend
08- HTML5
- CSS3
- React
- Next.js
- Angular
- Tailwind CSS
- Three.js
- React Three Fiber
Backend
03- FastAPI
- Node.js
- Express.js
AI / ML
10- LLMs
- RAG
- MCP
- PyTorch
- TensorFlow
- Keras
- scikit-learn
- NumPy
- pandas
- OpenCV
Databases
05- PostgreSQL
- MongoDB
- DynamoDB
- Pinecone
- Qdrant
Cloud & DevOps
03- AWS
- Docker
- Kubernetes
Tools & integrations
04- GitHub
- Socket.io
- Twilio
- ElevenLabs
06ARCHITECTURE STUDIES
Architecture
studies.
INTERACTIVE STUDIES
Conceptual workflows for retrieval, tool execution and computer vision.
GENERATIVE AI / RAG · ARCHITECTURE STUDY
Retrieval-augmented generation
A retrieval pipeline that supplies source material to a language model and returns a referenced response.
EmbeddingsRetrievalLLMs
NODE / 01
Question
Identify the request and retrieval requirements.
↳ Conceptual architecture studies. Project implementations are documented separately.
07CONTACT
Open to Software Engineer and AI/ML Engineer opportunities
Discuss
the work.
Open to software engineering and AI/ML roles. Send the problem, role or project context.
Contact detailsDEFINE THE PROBLEM. THEN DESIGN THE SYSTEM.