Lead Python Engineer with 1+ years in LLM Engineering & Multimodal AI
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Experienced Python Engineer specializing in Large Language Models (LLMs) and Artificial Intelligence. Led a cross-functional pod, delivering production-grade LLM pipelines and agentic backend services. Designed and deployed multi-agent reasoning workflows, adversarial prompt evaluation pipelines, and Supervised Fine-Tuning (SFT) pipelines. Proficient in developing microservice-based REST APIs with Python for LLM API-reasoning and agent interaction scenarios. Demonstrated expertise in multimodal AI, RAG systems, and MLOps, with a strong background in distributed systems and cloud deployments.
Presidency University
Bachelor of Technology · Computer Science & Engineering
September 1, 2020 – June 30, 2024
Kendriya Vidyalaya
12th Standard · PCMCs
March 1, 2019 – July 31, 2020
Turing
Python Engineer - LLM
August 1, 2024 – September 30, 2025
India
GenAI Multimodal Bot
June 24, 2026 – Present
Built a multimodal RAG agent using sqlite-vec and CLIP/BLIP for cross-modal retrieval. Integrated local LLMs (Ollama/Mistral) for offline, agentic QA over image and text corpora. Optimized latency using caching and efficient model loading strategies. Designed modular pipelines enabling extensible tool usage and reasoning workflows.
Mnemosyne – Agentic Multimodal Memory System
June 24, 2026 – Present
Built a fully offline agentic memory system enabling natural-language querying over complete screen history. Engineered a change-aware capture pipeline (OpenCV) that stores only meaningful visual updates, reducing storage footprint by 90% while preserving temporal continuity. Designed a multimodal indexing pipeline combining PaddleOCR-based text extraction with 1152-dim Google SigLIP embeddings for unified semantic representation of visual and textual content. Implemented hybrid retrieval using LanceDB, combining keyword search with vector similarity to surface contextually relevant historical states. Built an LLM-powered reasoning layer (Ollama) and Streamlit interface supporting conversational RAG, semantic search, time-travel replay, and screen activity analytics.
RAG QA System
June 24, 2026 – Present
Developed a production-grade RAG pipeline using FAISS for semantic document retrieval. Built FastAPI backend with SSE streaming for real-time, low-latency responses. Containerized deployment using Docker with scalable, cloud-ready architecture.
Cultural Fit Analysis
The candidate's personal projects demonstrate a strong passion for AI/ML, particularly in multimodal and agentic systems, which aligns well with innovative and research-oriented cultures. The experience leading a diverse team at Turing suggests adaptability and an ability to work effectively in varied technical environments. The breadth of skills and technologies indicates a continuous learning mindset.
Soft Skills & Operational Fit
The candidate's experience as a 'Pod Lead' at Turing indicates strong leadership, project management, and cross-functional collaboration skills. The detailed project descriptions suggest a proactive, problem-solving attitude and an ability to deliver complex technical solutions. The focus on optimizing latency, reducing storage, and building robust systems points to an operational mindset.