Saran Puneet

Data & AI Platform Engineer who ships from 1 TB/day streaming pipelines at Ola Electric to production RAG systems now running for enterprise and defense-sector clients at Webbverse AI. Builds where distributed systems (Kafka, Spark) meet applied LLM research (RAG, agentic workflows, vector search), backed by peer-reviewed NLP work (Springer, 2023). Engineers fast by pairing hands-on systems work with modern AI-augmented development tooling.

Founder & AI/Data Engineer at Webbverse AI (2026-02 – Present)

• Founded a specialized AI consultancy architecting enterprise-grade GenAI systems, integrating streaming data platforms (Kafka, Spark, Databricks) with multi-agent LLM workflows for B2B solutions.
• Engineering autonomous error-recovery pipelines and serverless deployments (NVIDIA NIM, LiteRT) to cut cloud and edge LLM inference latency
• Engaged with Indian Army (Western Command) at Webbverse AI to enhance surveillance capabilities through AI-driven systems; project scope and technical details are confidential.

Data Engineer at Kindrix Group (2024-06 – 2025-12)

• Created scalable AI data platforms for early-stage startups, integrating RAG pipelines (LangChain, Vector DBs, LLMs) for enterprise document parsing and retrieval.
• Owned end-to-end RAG system performance retrieval precision, source citation, and hallucination control across B2B client workloads, cutting document review cycles for enterprise clients.

DPE (Backend) at Sapper AI (Zyient) (2023-07 – 2023-11)

• Delivered automated ETL/ELT pipelines feeding LLM-powered RAG systems, reducing manual review effort for 3 enterprise clients.
• Cut unstructured data ingestion time by 40% by designing a robust data mapping framework that standardized heterogeneous JSON/text flows into structured formats for downstream AI processing.

SDE I at Ola Electric (2022-06 – 2023-07)

• Achieved 85% unit and integration test coverage while engineering telemetry ingestion pipelines processing 1 TB/day of streaming EV sensor data.
• Coordinated a cloud migration from on-premise to AWS/Azure, reducing data retrieval latency by 30% and infrastructure cost by 25%.

PRISM Developer (Research Internship) at Samsung (2020-07 – 2021-04)

• Engineered a novel NER data augmentation pipeline (Word2Vec, ELMo, BERT, Seq2Seq grammar correction, cosine-similarity sanity checker) for Samsung Bixby wake-word detection, improving F1 by 0.64 points (BERT) and 0.99 points (RoBERTa) on CoNLL-2003; published in the International Journal of Speech Technology (Springer, 2023).

Application Developer Intern at Monktree Education Pvt Ltd (2020-05 – 2020-07)

• Collaborated on front-end development of a financial mobile app (tax calculator), enhancing UI and functionality.

B.Tech in ECE – VIT University (2018 – 2022)

MA in Geography – VMOU University (2024 – 2026)