Engineer III

February 2025 till present • Bengaluru, IN

• Shipped a real-time NLP pipeline that live summarizes 1,000+ WebEx meetings weekly, cutting down manual meeting note taking.
• Built an NLP-powered Confluence search service (FastAPI, PGVector) indexing 2M+ pages, supporting sub-1500ms 95th-percentile search results latency.
• Tech stack: Python, FastAPI, Langchain, PGVector, Streamlit, Node.js, Puppeteer, WebEx JS SDK, OpenShift.

Associate Software Engineer

Jul 2023 till February 2025 • Bengaluru, IN

• Fine-tuned Mistral‑7B using LoRA adapters on 4070 Ti Super, baking company knowledge into model.
• Deployed the fine-tuned model on a 4x1080 Ti GPU cluster via vLLM; engineered dynamic batching and employed mixed-precision inference for sub-750ms 95th-percentile Time-To-First-Token.
• Led development of "Beacon", an NER-based print-request automation service integrating with emails and natural language interfaces, leveraging a custom fine-tuned ModernBERT model.
• Built AskDB, a natural language to insights platform, used by 300+ internal users to query company DBs and auto-generate visual insights.
• Tech stack: vLLM, Python, PyTorch, LoRA, SFT, Mistral, Llama, Ollama, SQL, FastAPI.

Data Scientist, Intern

Jan 2023 till Jul 2023 • Bengaluru, IN

• Created a product mockup pipeline using ImageMagick and Node.js, achieving cost-effective design automation.
• Tech stack: Node.js, ImageMagick, Angular.

Fullstack Developer, Intern

Jul 2022 till Dec 2022 • Hyderabad, IN

• Migrated Flask monolith to microservices (Python, Docker, Kubernetes), optimizing AutoML job triggering and reducing job-launch latency by 40%.
• Tech stack: Python, Flask, Docker, Kubernetes.