Project
Adaptive LLM Inference Engine
A resource-constrained system for choosing model size, quantization level, and execution settings based on available device capacity.
Model selection
Quantization
Resource-aware AI
Backend AI, Built Right
Santosh M builds backend and AI systems for real products.
Focused on Backend Development and AI/ML Engineering for tech enthusiasts and potential collaborators. Python, FastAPI, APIs, databases, LLM applications, and resource-aware AI systems shape the work.
Current focus
Backend + AI/ML
Python, FastAPI, APIs, databases, model integration, and deployment-friendly engineering.
Exploration
LLM systems
RAG, LangChain, LangGraph, prompt engineering, local models, quantization, and resource-aware inference.
Personal mission statement
Build dependable systems, then make them usable.
Santosh M focuses on backend development and AI/ML engineering with a practical bias: shipping APIs, integrating models, and shaping tools that remain maintainable when the prototype ends.
01
Backend Development
FastAPI services, database-backed workflows, API design, and integration work that keeps product logic clean.
02
AI/ML Engineering
LLM applications, RAG, LangChain, LangGraph, AutoML, and resource-aware model selection for practical use cases.
03
Prompt Engineering
Prompt design, evaluation, and pipeline thinking for AI features that feel useful rather than fragile.
Detailed project deep dives
Selected practical work and learning artifacts.
A small set of projects that show the bridge between backend systems, AI engineering, and product thinking.
Academic project
Adaptive LLM Inference Engine
Backend development
API Service Layer
Product development
CoAI.Pro Workstream
AI integrations
Prompt Engineering Experiments
Learning project
Model Optimization Notes
Systems design
ML Integration Prototype
Skills and tools used
The stack stays lean: build, ship, refine.
Santosh M works across the product surface and the service layer, with enough breadth to move from idea to implementation without losing sight of maintainability.
Backend
Python, FastAPI, APIs
Service design, request handling, validation, and clean integration points.
Data
Databases, retrieval, pipelines
Data modeling, query-aware workflows, and retrieval support for AI use cases.
AI
LLMs, RAG, LangChain
Application patterns for search, generation, evaluation, and assistant flows.
Research
Local LLMs, quantization
Model optimization, resource-aware execution, and deployment-sensitive tradeoffs.
Case studies of previous work
Three contexts where systems thinking mattered.
These examples focus on the shape of the work: what was being solved, how the stack was approached, and where AI fit into the system.
Project
CoAI.Pro backend work
Backend development, AI integrations, prompt engineering, deployment plumbing, and the pipelines that support website generation.
APIs
Prompting
Deployment
Project
Learning systems for modern LLM apps
Exploration around RAG, LangGraph, local models, and the practical tradeoffs involved in building applications that stay responsive.
RAG
LangGraph
Optimization
Technical blog articles
Notes on backend engineering and applied AI.
A small editorial feed for the ideas, experiments, and systems questions that shape the work.
Backend
Designing FastAPI services for clarity
How the routing, schema, and service boundaries come together in small but maintainable systems.
AI
Local models and resource-aware AI
A practical view of model size, quantization, and execution constraints for edge and lightweight deployments.
Product
Prompt engineering inside product pipelines
How prompt design, feedback loops, and deployment decisions influence the stability of AI features.
About
The work is about dependable systems and clear communication.
Santosh M is building a direction around backend development and AI/ML engineering. The focus stays on practical product work: APIs, data flow, model integration, and the glue that lets tools be shipped and maintained.
Current learning spans LLM applications, RAG, LangChain, LangGraph, AutoML, generative AI, local models, quantization, and resource-aware inference. The goal is to contribute to teams that value technical clarity and thoughtful implementation.
Focus
Backend first, AI aware.
A steady mix of implementation, experimentation, and learning — aimed at collaborators who want thoughtful execution rather than surface-level features.
Python
FastAPI
LLMs
Databases
Get in touch
Send a message about a project, collaboration, or learning opportunity.
This is the only direct action on the site. Use it to start a conversation about backend systems, AI features, or a shared build.
Newsletter sign-up
A simple note when there is something worth sharing.
No extra capture funnels. This section exists as a light editorial placeholder for occasional updates, experiments, or write-ups.
Follow along for technical notes.
Updates are occasional and focused on backend work, AI systems, and practical learning.