Headline

AI Engineer | Automating Workflows with LLM Systems | Backend & Infrastructure

About

I build AI systems that actually get used in production. Currently working as an AI Engineer at Wingify, where I design and ship LLM-powered systems across customer support, sales intelligence, and workflow automation. What I’ve worked on: • Built a GenAI support system handling 5K+ monthly queries with ~92% accuracy • Reduced SDR research time from 3 hours → 4 minutes using AI pipelines • Improved outreach response rates by ~40% through AI-driven insights • Built 25+ internal AI tools for debugging, automation, and workflows • Set up LLM observability (Langfuse), reducing costs by ~20% Previously at Newgen Software: • Built RAG pipelines for document intelligence systems • Created a 6,000-query dataset using Mistral-7B • Fine-tuned embedding models (NDCG@10: 45.52 → 54.31) • Improved LLM retrieval performance by ~20% I enjoy working at the intersection of backend systems and AI especially around RAG pipelines, LLM fine-tuning, and building scalable GenAI products. Currently focused on building reliable, production-grade AI systems not just demos. Reach out to me at: ai.ashishnehra@gmail.com

Location

Greater Jaipur Area

AI stack

claude code

cursor

notebooklm

ollama

ChatGPT Images

Firecrawl

Work experience

Software Developer GenAI at Wingify, Jul 2024 - Present · 2 yrs 2 mos, Delhi, India, - Architected a GenAI-powered customer support system handling 5K+ monthly queries with ~92% accuracy using RAG pipelines and AI agents - Built an AI-driven sales intelligence platform reducing SDR research time by 97% (3 hrs → 4 mins) and increasing reply rates by 40% - Developed 25+ internal AI tools for debugging, automation, and workflow execution across support pipelines - Designed a personalized deck generation system reducing creation time from 1 day → 30 minutes - Implemented LLM observability using Langfuse, reducing operational costs by 20% - Built automated campaign and landing page generation workflows reducing creation time by ~98% - Developed AI-powered outreach systems using web scraping + LLMs to generate high-conversion messaging

Data Science Intern at Newgen Software, Jan 2024 - Jun 2024 · 6 mos, Gurugram, Haryana, India, - Developed RAG (Retrieval-Augmented Generation) pipelines, including data splitting techniques, pre-retrieval techniques, and advanced retrieval methods - Data-Generation: Created a 6000 query-corpus dataset using 300 documents with Mistral-7B LLM. - Embedding Finetuning: Fine-tuned the GTE-large embedding model using the above dataset, improving NDCG@10 from 45.52 to 54.31. - Instruction fine-tuned Mistral-7B LLM for better response generation, enhancing retrieval performance by 20%.

AI Software Developer Intern at AmberFlux EdgeAI Private Limited, Jun 2023 - Aug 2023 · 3 mos, Hyderabad, Telangana, India, - Engineered two innovative products: PDF-centered Q&A system and Audio Transcript Analyzer. - Applied transformer-based models to speech to text conversion and leveraged the generated transcripts to generate insights and create Q&A functionalities. - Implemented chained conversation for enhanced responses using Langchain framework.

ML Pioneer at MarkovML, May 2023 - Jun 2023 · 2 mos, Bengaluru, Karnataka, India, - Initiated the development of an advanced Dialogue-based system that incorporates emotional context to enhance responses using the MarkovML platform. - Implemented emotion recognition and speech-to-text using 2D neural networks and transformers.

ML Contributor at Omdena, Mar 2023 - Apr 2023 · 2 mos, Remote, - Collaborated on developing a Real Estate Platform application - Implemented a Machine Learning model for impactful real estate recommendations, showcasing skills in data analysis and modeling.

Education

Bachelor of Technology - BTech, Information Technology, ABV-Indian Institute of Information Technology and Management

Public links

github: https://github.com/Ashish-Nehra