AI/ML
→
Axon
AI-powered second brain — a deployed RAG knowledge assistant
FastAPIChromaDBHuggingFaceGroqGemini+2
System Initialization
Gouribhatla Sai Vidith
Initializing neural pathways and establishing backend connections...
Click Anywhere to Enter
Available to Build
Sai Vidith
I design and build intelligent systems that scale. From robust backend architectures and cloud deployments to cutting-edge machine learning pipelines, I focus on bringing complex engineering concepts to life.
6-8GPU Cluster Nodes
100+Docs Automated
40+Hrs/Month Saved
200+Community Led
// ABOUT_ME
System Dashboard
SV
Sai Vidith
AI/ML & Backend Engineer
VNR Vignana Jyothi Institute of Engineering and Technology
Penultimate-year CSE (Data Science) student building production systems across AI/ML, backend, and infrastructure — from an 8-GPU compute cluster I run myself to RAG pipelines and a 17-year document pipeline built at INCOIS.
PythonFastAPIRayYOLOv8Next.jsDocker
8.87
/ 10
CGPA
200+
members
Community Led
40+
hrs / month
Manual Effort Cut
160+
LeetCode
Problems Solved
Currently Learning
- →Kubernetes & container orchestration at scale
- →LLM fine-tuning & RLHF workflows
- →Diffusion-based video generation (ComfyUI)
- →Distributed systems at cluster scale
leadership — achievements.log
→
Tech LeadVJ DataQuesters· 2024 – Present
Led technical initiatives for a 200+ member community (50+ technical); shipped production features to the club platform and mentored junior members on engineering best practices.
→
Runner-upTechnovista Hackathon· 2024
Delivered a working AI-based system under time constraints, demonstrating end-to-end problem solving and team execution in a competitive setting.
→
Rank 47 / Top 100Competitive Programming Contest
Solved 160+ problems across arrays, graphs, dynamic programming, and algorithms on LeetCode.
🛡️
Design for failure
Every system will fail. Design with circuit breakers, retry logic, and graceful degradation from day one.
⚡
Cache aggressively, invalidate carefully
Cache at the right layer. Redis for hot data, CDN for static, in-memory for computed values. TTL > manual invalidation.
📐
APIs are contracts
API design is product design. Version from v1, document everything, never break backwards compatibility in production.
📊
Measure, then optimize
Premature optimization is the root of all evil. Profile first, then optimize the actual bottleneck with data.
Selected Work
Projects
AI/ML
→
AI Business Tool Kit
MCP multi-agent platform — 5 AI microservices in production
Node.jsMCPDockerRedisMongoDB+2
Infrastructure
→
VJ GPU Cluster
Distributed Ray compute cluster across 6–8 physical nodes
RayPythonDockerLinuxFastAPI+2
AI/ML
→
Emergency Vehicle Detection
Real-time emergency vehicle detection at 25+ FPS with YOLOv8
YOLOv8PyTorchOpenCVPython
More Work
Research
ESG Greenwashing Detection
IEEE — submittedResearch applying Gradient Boosting to detect ESG greenwashing signals across 835 firms, using SHAP for interpretability and a Streamlit app to explore predictions.
835 firms analysedF1 = 0.97SHAP-based interpretability
Pythonscikit-learnSHAPStreamlit
Architecture
System Design
Request Lifecycle
User
Request
API Gateway
Rate limit + Auth
Cache
Redis / CDN
Service
Business logic
Database
PostgreSQL
Response
Serialized
Caching
Redis for hot data (TTL: 30s). CDN for static assets. Cache-aside with write-through for consistency.
Latency
p50 <50ms, p99 <200ms. Cache hit <10ms. DB query <80ms. Budget: 20% network, 30% compute, 50% I/O.
Scaling
Horizontal via stateless services. DB read replicas. Event-driven async for compute. Circuit breakers at boundaries.
Design for failure
Every system will fail. Design with circuit breakers, retry logic, and graceful degradation from day one.
Cache aggressively, invalidate carefully
Cache at the right layer. Redis for hot data, CDN for static, in-memory for computed values. TTL > manual invalidation.
APIs are contracts
API design is product design. Version from v1, document everything, never break backwards compatibility in production.
Measure, then optimize
Premature optimization is the root of all evil. Profile first, then optimize the actual bottleneck with data.
// SKILLS
Tech Arsenal
Tools and technologies I use to build production-grade systems.
💻
Languages
PythonJavaScriptTypeScriptSQLBashJava
🧠
AI / ML & Data
PyTorchHuggingFaceYOLOv8RAG / ChromaDBNumPyPandasData Visualisation
⚙️
Backend & APIs
FastAPINode.jsExpress.jsREST APIsWebSocketsMCP
🗄️
Databases
PostgreSQLMongoDBMySQLRedisChromaDB
☁️
Cloud & DevOps
DockerKubernetes (basics)LinuxNginxAWS (basics)GCP
🛠️
Tools
GitGitHubJIRASplunk (basics)
// CERTIFICATIONS
🏆
AI Fundamentals
IBM SkillsBuild / Cisco
🏆
Prompt Design in Vertex AI
Google Cloud
🏆
SQL
HackerRank
// EXPERIENCE
Journey Log
💼
Fullstack Intern
Trovofi Pvt Ltd
Oct 2025 – Dec 2025trovofi.in
Built and shipped the frontend and backend for www.trovofi.in over a 3-month internship, including automated email workflows and database-backed storage for user data.
Frontend + backend of trovofi.inEmail automation with NodemailerFirebase for data storage
💼
AI/ML Project Intern
INCOIS (Indian National Centre for Ocean Information Services)
Mar 2025 – May 2025Hyderabad, India
Designed and implemented an end-to-end Python data pipeline to ingest, process, and structure 17 years of domain-specific documents (2008–2025) — ETL across 100+ large-scale PDFs with OCR fallback for scanned documents. Automated document workflows to eliminate 40+ hours/month of manual effort, and maintained the pipeline end-to-end, diagnosing and fixing format-specific extraction failures.
17 years of documents processed100+ PDFs, OCR fallback40+ hrs/month automatedWorked directly with domain scientists
⚡
Tech Lead
VJ DataQuesters
2024 – PresentHyderabad, India
Led technical initiatives for a 200+ member community (50+ technical); shipped production features to the club platform and mentored junior members on engineering best practices.
200+ member communityProduction features shippedMentored junior members
🏆
Runner-up — Technovista Hackathon
Technovista Hackathon
2024
Delivered a working AI-based system under time constraints, demonstrating end-to-end problem solving and team execution in a competitive setting.
🏆
Rank 47 / Top 100 — LeetCode
Competitive Programming Contest
Solved 160+ problems across arrays, graphs, dynamic programming, and algorithms on LeetCode.
160+ problems solved
🎓
B.Tech, Computer Science and Engineering (Data Science)
VNR Vignana Jyothi Institute of Engineering and Technology
2023 – 2027Hyderabad, India
CGPA: 8.87 / 10
CGPA 8.87
// RESUME
The Full Record
Everything above, condensed to one page — or grab the PDF.
📄
Sai Vidith — Resume
SaiVidith_Resume.pdf
LocationHyderabad, India
Summary
Penultimate-year CSE (Data Science) student building production systems across AI/ML, backend, and infrastructure — from an 8-GPU compute cluster I run myself to RAG pipelines and a 17-year document pipeline built at INCOIS.
Education
VNR Vignana Jyothi Institute of Engineering and Technology
B.Tech, Computer Science and Engineering (Data Science)
2023 – 2027
CGPA 8.87
Experience
Fullstack InternTrovofi Pvt Ltd
Oct 2025 – Dec 2025Built and shipped the frontend and backend for www.trovofi.in over a 3-month internship, including automated email workflows and database-backed storage for user data.
AI/ML Project InternINCOIS (Indian National Centre for Ocean Information Services)
Mar 2025 – May 2025Designed and implemented an end-to-end Python data pipeline to ingest, process, and structure 17 years of domain-specific documents (2008–2025) — ETL across 100+ large-scale PDFs with OCR fallback for scanned documents. Automated document workflows to eliminate 40+ hours/month of manual effort, and maintained the pipeline end-to-end, diagnosing and fixing format-specific extraction failures.
Skills
LanguagesPython, JavaScript, TypeScript, SQL, Bash, Java
AI / ML & DataPyTorch, HuggingFace, YOLOv8, RAG / ChromaDB, NumPy, Pandas, Data Visualisation
Backend & APIsFastAPI, Node.js, Express.js, REST APIs, WebSockets, MCP
DatabasesPostgreSQL, MongoDB, MySQL, Redis, ChromaDB
Cloud & DevOpsDocker, Kubernetes (basics), Linux, Nginx, AWS (basics), GCP
ToolsGit, GitHub, JIRA, Splunk (basics)
Certifications
AI Fundamentals · IBM SkillsBuild / CiscoPrompt Design in Vertex AI · Google CloudSQL · HackerRank
// CONTACT
Let's Connect
Open to internships, collaborations, and interesting engineering problems.
Available for opportunities
Find me online
// QUICK_NOTE
I respond quickly. Whether it's about a project, a collaboration, or a role — I'm genuinely interested. Drop a message.
Send a message
© 2025 Gouribhatla Sai Vidith · Built with Next.js + Framer Motion · Hosted on saividith.tech