App LogoApp name
SJ

Shivam jha

Open to work2 years experience

AI Software Engineer

Dehradun, IndiaTarget Roles: Backend Engineer • Full-Stack Developer • Frontend Specialist

AI Software Engineer specializing in GenAI, RAG, and scalable backend systems.

GitHubLinkedIn
Resume
Email

Standing Rank

Rank Not Available

Developer Badges

Badges Pending

Skills & Technologies

50 skills
langchainragpineconechromadbfaissloraqlorapeftopenaihuggingfaceprompt engineeringllm orchestrationai agentsclaude apipytorchtransformerspythonjavajavascripttypescriptdartc++sqlbashreactnext.jsreduxtailwind cssjestreact nativeflutternode.jsexpress.jsrest apismicroservicesredislaravelawsdockerci/cdlinuxmongodbpostgresqlmysqlgitjirapostmansocket.iok6streamlit

Work Experience

AI Software Engineer

Academically Pvt Ltd

Jun 2024 - Present Dehradun, India
  • Mentored 5+ developers on system design, AI integration patterns, and agile delivery, improving team velocity by 25%
  • Developed automated course access control using scheduled jobs and event-driven architecture, processing 2,000+ student accounts monthly with 100% accuracy

Software Engineer Intern

ColoredCow

Jan 2024 - Jun 2024 Tehri-Garhwal, India
  • Engineered RESTful APIs using Laravel/MySQL within microservices architecture for enterprise tree planting system (Bank of Baroda, Phillips), handling 50K+ records
  • Optimized MongoDB queries with compound indexing, reducing retrieval time by 20% for 100K+ records; contributed to 25% faster delivery via Agile sprint planning

Projects

LinkedIn Post Automation – GenAI System

LangChainClaude AIRAGVector DBPython
  • Built end-to-end GenAI content automation using LangChain agents, Claude (Anthropic), and OpenAI function calling with custom prompt chaining and few-shot prompting for high-engagement content generation
  • Implemented production RAG pipeline with vector database (Pinecone/ChromaDB), recursive text chunking, and embedding models (OpenAI Ada-002, HuggingFace sentence-transformers) for semantic retrieval, improving relevance by 60%
  • Designed semantic search via vector embeddings with cosine similarity, top-k retrieval, and dynamic context injection into LLM prompts; built guardrails for output validation, reducing manual effort by 80%

PharmaHacker Assessment Platform

ReactReduxNode.jsMongoDBRedisAWS SQSFlutter
  • Led team building high-concurrency testing platform serving 23,000+ pharmacists with 10,000+ simultaneous users, achieving 99.9% uptime during peak load
  • Architected backend with Redis caching and MongoDB connection pooling, reducing DB latency by 70%; implemented AWS SQS with 8 queues + DLQ for zero-loss batch processing

CRM System

ReactContext APITailwindNode.js
  • Developed Course Extension Dashboard with RBAC using Node.js/React; automated referral workflows via event-driven architecture, reducing manual processing by 100%

Intelligent Document Q&A System

PythonLangChainOpenAIFAISSFine-Tuning (LoRA/QLoRA)Streamlit
  • Built RAG-powered Q&A chatbot using LangChain RetrievalQA chain – ingests PDFs via document loaders, applies recursive character splitting, generates vector embeddings (OpenAI Ada-002), and indexes in FAISS vector store for semantic retrieval with conversational memory
  • Performed fine-tuning with LoRA/QLoRA on open-source LLMs (LLaMA, Mistral) using HuggingFace Transformers and PyTorch for domain adaptation, achieving 15% accuracy improvement on custom evaluation benchmarks
  • Deployed via Streamlit with real-time token streaming, configurable temperature/top-k/top-p inference parameters, and model evaluation metrics (BLEU, ROUGE, cosine similarity) for response quality monitoring

Leaderboard Standings

Leaderboard Position Pending

Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.

Assessment Highlights

Assessments Not Completed

Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.

AI Collaboration Score

AI Collaboration Score Pending

Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.

Role Compatibility Profile

Role Compatibility Analysis Pending

Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.

Achievements

1st Place – National Hackathon (Hackmol-5.0)

1st Place – National Hackathon (Hackmol-5.0) at NIT Jalandhar

Runner-up – Uhackathon 4.0

Runner-up – Uhackathon 4.0

About Details

Professional Bio

AI Software Engineer with expertise in GenAI, RAG pipelines, and scalable backend systems. Proven track record in building high-concurrency platforms and optimizing system performance. Experienced in full-stack development and AI integration.

Bachelor of Technology

THDC-IHET (UTU) (2020 - 2024)

Languages: English, Hindi
Email: Locked