I build LLM applications that retrieve knowledge, use tools, and connect to real workflows.
Python at the core. FastAPI at the boundary. AWS in the cloud.
Profile · Projects · Stack · Experience
I'm Aditya, an AI/ML & Cloud Engineer based in Hyderabad, India, currently working as a Junior AI Cloud Engineer. I build AI-powered features, authenticated serverless APIs, and enterprise knowledge-source integrations. My earlier work includes document-grounded assistants and offline-capable operations dashboards.
My work spans the full path from chunking and embeddings to agent orchestration, streaming APIs, and deployment troubleshooting. I also build in the open: local agents, text-to-app experiments, and accessible model fine-tuning.
| RETRIEVE Document ingestion, semantic search, and answers grounded in source material. |
ORCHESTRATE Agent workflows, tool calling, structured outputs, and streaming responses. |
SHIP Serverless APIs, authenticated integrations, infrastructure as code, and runtime monitoring. |
Under the hood — the engineering decisions I care about
| Boundary | Technical focus |
|---|---|
| Knowledge → context | Chunking, embeddings, similarity search, and preserving source metadata for citations. |
| Model → application | Tool calling, structured output parsing, provider integration, and streamed responses. |
| Integration → cloud | OAuth connection flows, Cognito sessions, authenticated APIs, and Lambda/SQS/S3 ingestion. |
| Code → operation | SAM/CloudFormation deployment troubleshooting, CloudWatch runtime monitoring, and release checks. |
| Model → adaptation | PEFT/LoRA experiments for small models, with CPU-based local training and chat. |
Local-first agents → multi-agent generation → model customization
Three complementary directions in my open-source work: running assistants locally, coordinating agents to build applications, and adapting small models on consumer hardware.
| Project | What I built | Engineering focus |
|---|---|---|
| ALD-01 | Local AI agent system with 5 specialized agents, 10+ LLM integrations, provider failover, and a streaming dashboard. Published on npm and PyPI. | Python FastAPI Ollama LangChain |
| CodeGenesis | Multi-agent text-to-app platform that generates full-stack application code from natural-language prompts. | LangChain LangGraph OpenRouter FastAPI |
| FineTuneLite | Local fine-tuning studio for small language models, with CPU-based PEFT/LoRA training and chat. | Python PEFT LoRA Transformers |
| Intelligent Document System | Document Q&A for consulting and staffing workflows, with source citations and confidence scores. | RAG Embeddings Document Q&A |
| DYSC-Agent | Security analysis CLI combining static analysis, AI-assisted fix suggestions, and an interactive tool-calling REPL. | Node.js CLI Static analysis |
| SafeBite | Food-analysis platform with Gemini nutrition insights, OCR label scanning, Firebase authentication, and real-time sync. | Flask Node.js MongoDB Firebase |
| Layer | Tools & concepts |
|---|---|
| Intelligence | LangChain · LangGraph · CrewAI · RAG · MCP · tool calling · structured outputs · RAG evaluation |
| Models & data science | PyTorch · scikit-learn · Hugging Face Transformers · SentenceTransformers · PEFT/LoRA · Ollama · Pandas · NumPy |
| Services & data | Python · FastAPI · Flask · Node.js · REST · WebSockets · PostgreSQL · MongoDB · SQL Server · FAISS · ChromaDB |
| AWS | Lambda · S3 · EC2 · Cognito · Amplify · SQS · CloudWatch · SAM · CloudFormation |
| Delivery & security | GitHub Actions · CI/CD · Docker · infrastructure as code · OAuth 2.0 · JWT · RBAC |
| Languages & tooling | Java · JavaScript · Git · Postman · npm · PyPI · GCP & Kubernetes fundamentals |
Junior AI Cloud Engineer
Sep 2026 → Present
- Deploy AI-powered features on AWS and implement Cognito sessions, OAuth integrations, and authenticated serverless APIs.
- Implement Slack connection flows and support history ingestion with Lambda, SQS, and S3.
- Troubleshoot SAM/CloudFormation deployments, monitor runtime health in CloudWatch, and maintain release checks and technical documentation.
AI Engineer Intern
Jan 2026 → Jun 2026
- Built a RAG chatbot over 100+ enterprise documents using LangChain, Ollama, FAISS/ChromaDB, and FastAPI, with source-cited answers.
- Developed document-processing workflows and Talent Radar, a candidate-matching dashboard.
- Integrated LLM providers with structured output parsing and streaming responses.
Full Stack Developer Intern
Jul 2025 → Sep 2025
- Built an operations dashboard with Node.js and MongoDB, supporting offline field workflows through Service Workers.
- Implemented multi-user authentication, RBAC, and real-time data synchronization.
Earlier experience
Software Engineering Intern · Apr 2025 → Jun 2025
Built enterprise HRMS APIs with VB.NET and SQL Server; worked on query optimization and bulk CSV/Excel processing.
Engineering priorities
- Grounded answers: preserve source context and make citations useful.
- Explicit interfaces: use structured outputs and clear tool contracts.
- Secure integrations: verify authentication and scope access to the workflow.
- Operable services: make failures visible through logs, monitoring, and release checks.
B.Tech · Computer Science Engineering
ICFAI Foundation for Higher Education, Hyderabad · Jul 2022 – Sep 2026
- Google AI Essentials · 2026
- AWS Machine Learning Foundations · 2025
- AWS Cloud 101 · 2025
- Introduction to Kubernetes (LFS158) · Linux Foundation · 2025
- Smart India Hackathon 2024 · National Certificate
- ISRO Space Technology Training · 2024
Let's build useful AI, from retrieval to deployment.
LinkedIn · Portfolio · Let's build something useful