Backend + AI engineer in San Francisco, California β available to work.
3+ years of professional experience Β· MS CS @ CU Boulder Β· Focus: scalable systems + AI
I build production systems at the intersection of backend engineering and AI β from RAG platforms serving 2M+ users to LLM agents, observability pipelines and cloud-native services. AI-native builder: I use AI technologies end-to-end to design, ship and iterate on anything faster β day-to-day coding with Claude and Cursor.
Roles: Cisco Extern (CU Boulder) Β· finish'd Β· Kore.ai Β· Express Stores
π’ Cisco Systems β AI Engineer Extern Β |Β CU Boulder Γ Cisco Β |Β 2025
- LLM-powered co-pilot for network engineers β configs, intents and topologies via LangChain and MCP
- Built agent workflows that help engineers navigate complex network environments faster
LangChain MCP Python LLMs
π’ finish'd, Inc. β Software Engineer Intern Β |Β Denver, CO Β |Β May 2025 β Jun 2025
- Built a scheduled change detection pipeline across 60 pages on GCP using LangChain, Cloud Run, Scheduler and Terraform β 80% less manual tracking saving 6 hrs/week
- Automated release publishing via LLM parsing and Firestore validation β 50% less curation, 90% fewer feed mismatches
LangChain GCP Terraform Firestore Cloud Run
π’ Kore.ai β Software Engineer Β |Β Hyderabad, India Β |Β Jan 2024 β Jul 2024
- Built production RAG search service using LangChain, LlamaIndex, Elasticsearch and MongoDB β 2M+ users, 90% better answer quality
- Implemented hybrid BM25 + dense vector retrieval with analyzers, boosts and synonyms β 25% better top-k hit rate
- Fine-tuned embedding models and trained rerankers using Hugging Face Transformers β 40% better retrieval accuracy
LangChain LlamaIndex Elasticsearch MongoDB Hugging Face
π’ Kore.ai β Associate Software Engineer Β |Β Hyderabad, India Β |Β May 2022 β Dec 2023
- Custom New Relic instrumentation across Python, Node.js, RabbitMQ, Redis and MongoDB β 85% faster debugging (2 hrs β 20 mins)
- Built distributed Node.js ETL pipeline with OAuth2.0 across Confluence, SharePoint, Zendesk and ServiceNow β 90% more searchable knowledge
- Automated GPU scaling on AWS using boto3 and Jenkins β $50K/yr saved
- Designed and led intern onboarding program β 10% team productivity increase
New Relic Node.js OAuth2.0 AWS Jenkins RabbitMQ Redis
π’ Express Stores β Full Stack Software Engineer Β |Β Gurgaon, India Β |Β Jul 2021 β Apr 2022
- Built Django REST APIs with S3, Paytm and Zoho integrations and optimized PostgreSQL queries β 50% faster data retrieval
- Built POS screens in React and React Native with pagination, caching and skeleton loaders β 30% better dev velocity, 15% higher conversion
Django PostgreSQL React React Native S3
| π | π | π° |
|---|---|---|
| 2M+ users served by production RAG system | 85% faster incident debugging | $50K/yr saved via GPU autoscaling |
| 90% improvement in answer quality | 240+ LeetCode problems solved | 3+ years professional experience |
|
π€ Cisco AI Engineer Assistant Extern Β· Cisco Systems Γ CU Boulder LLM-powered co-pilot helping network engineers navigate configs, intents and topologies using LangChain and MCP.
|
Data Mining Course Project Full pipeline on NHTSA FARS 2023 β ML models, SHAP, association rules and a live Streamlit crash hotspot explorer.
|
|
Real-time crypto sentiment platform. Ingests Reddit and news, scores with BERT and VADER, streams live WebSocket charts and explains sentiment shifts with OpenAI. 5 backend services, 10 coins, CI pipelines per service. |
Automates event creation by extracting details from images and audio using Tesseract OCR and Whisper, then syncs directly to Google Calendar. Deployed with Docker and Kubernetes. 90% fewer manual errors, saves 3+ hrs/week. |
|
Extracts pill imprints from photos using AWS Rekognition, looks up FDA drug data and answers safety questions via OpenAI. Includes LASA error detection and Redis caching. |
Multi-server MCP agent with Claude and Gemini clients connected to a custom arXiv research server. Searches papers, extracts metadata and generates query-specific digests over stdio and HTTP transports. |
| Degree | Institution | Year |
|---|---|---|
| MS in Computer Science | University of Colorado Boulder | 2024 β 2026 |
| BE in Electronics and Communication Engineering | BITS Pilani | 2018 β 2022 |