Akansha Pal

Akansha Pal

Gen AI Engineer
Noida , Uttar Pradesh , India
Full-time On-site

About Me

AI Engineer with 3+ years of experience in Generative AI and Computer Vision.Built projects like RAG Chatbot, Multi-Agent RAG, Object detection ,image classification, OCRs.Experience in REST APIs, RAG systems, and LangCh…

Experience

Gen AI Engineer

WESEE-Ministry of Defence, New Delhi, India · New Delhi , Delhi , India
Nov 2024 - Present · 1 year 7 months

Built a production RAG-powered Q&A chatbot using LangChain for data ingestion, chunking, and embedding storage across ChromaDB and Milvus — enabling semantic search over large document corpora., Integrated vLLM for high-throughput streaming inference, significantly reducing response latency in real-time chatbot interactions., Implemented Speech-to-Text using faster-Whisper for high-accuracy multilingual transcription from live audio, audio files, and video files., Built a multilingual translation pipeline with NLLB-200 and CTranslate2, supporting 20+ languages with optimized inference for near real-time throughput., Designed a Neural Text-to-Speech system using Kokoro TTS with multiple speaker voices and speech speed control.

Gen AI Engineer

WESEE-Ministry of Defence,(Payroll-Simple Logtech Private Limited)
Nov 2024 - Present · 1 year 8 months

Built a production RAG-powered Q&A chatbot using LangChain for data ingestion, chunking, and embedding storage across ChromaDB and Milvus, enabling semantic search over large document corpora.
Integrated vLLM for high-throughput streaming inference, significantly reducing response latency in real-time chatbot interactions.
Implemented Speech-to-Text using faster-Whisper for high-accuracy multilingual transcription from live audio, audio files, and video files.
Built a multilingual translation pipeline with NLLB-200 and CTranslate2, supporting 20+ languages with optimized inference for near real-time throughput.
Designed a Neural Text-to-Speech system using Kokoro TTS with multiple speaker voices and speech speed control.

Computer Vision Engineer

Orama solutions LLP · Delhi , Delhi , India
Aug 2023 - Aug 2024 · 1 year

Developed and deployed computer vision models for applications such as object detection, image classification, optical character recognition (OCR), and defect detection., Optimized deep learning models using TensorFlow and PyTorch, enhancing performance and accuracy by 15%., Conducted exploratory data analysis (EDA) and applied data preprocessing techniques, including tokenization and bag of words, while implementing large language models (LLMs) like Llama., Enhanced model training efficiency through data augmentation and annotation., Developed and deployed RESTful APIs for YOLO and OCR algorithms using Flask API and FastAPI, which facilitated seamless integration.

Computer Vision Engineer

Orama Solutions LLP, New Delhi, India
Aug 2023 - Aug 2024 · 1 year

Developed and deployed computer vision models for applications such as object detection, image classification, optical character recognition (OCR), and defect detection.
Optimized deep learning models using TensorFlow and PyTorch, enhancing performance and accuracy by 15%.
Conducted exploratory data analysis (EDA) and applied data preprocessing techniques, including tokenization and bag of words, while implementing large language models (LLMs) like Llama.
Enhanced model training efficiency through data augmentation and annotation.
Developed and deployed RESTful APIs for YOLO and OCR algorithms using Flask API and FastAPI, which facilitated seamless integration.

Computer Vision Engineer

Nex-G Exuberant Solutions Pvt. Ltd., Noida, India · Greater Noida , Uttar Pradesh , India
Sep 2022 - Aug 2023 · 10 months

Developed and deployed computer vision models for applications such as object detection, image classification., Researched and implemented computer vision pipelines, enhancing model accuracy., Maintained comprehensive technical documentation, streamlining knowledge transfer and onboarding, resulting in a faster integration time., Annotated datasets and visualized results

Computer Vision Engineer

Nex-G Exuberant Solutions Pvt. Ltd., Noida, India
Sep 2022 - Aug 2023 · 11 months

Developed and deployed computer vision models for applications such as object detection and image classification.
Researched and implemented computer vision pipelines, enhancing model accuracy.
Maintained comprehensive technical documentation, streamlining knowledge transfer and onboarding, resulting in a faster integration time.
Annotated datasets and visualized results.

Skills

MySQL Python Flask Keras OpenCV TensorFlow Transformers Docker Elasticsearch LLMs Prompt Engineering RAG Pipelines Agentic AI MCP Architecture A2A LangChain LangGraph Hugging Face LlamaIndex OpenAI Ollama vLLM VLMs Fine-tuning (LoRA QLoRA) Multimodal AI Automatic1111 OpenTelemetry PyTorch YOLOv5 YOLOv8 PaddleOCR NLP OCR Object Detection Image Classification Data Augmentation ChromaDB Milvus Pinecone FastAPI REST APIs WebSockets Git Fine-tuning LoRA QLoRA
Report this Profile?