Portfolio — 2026Coimbatore, IN · 11.01° N, 76.95° EOpen to opportunities

PrasannaB.AI / ML Engineer

I build RAG assistants, AI video & image editors and autonomous agents that run in production — and every Sunday I teach robotics & AI to students from Class 2 upward.

CurrentlyAI / ML EngineerCreative Bees · since Sept 2026Robotics Eng. → Python Dev. → AI / ML
Engineering, robotics → AI
3+yrs
Production inference latency
<200ms
Face liveness accuracy
99.2%
Students & engineers mentored
150+
Scroll
01 — About

From robotics to applied intelligence.

A three-year technical trajectory: two years in robotics engineering, one year building backend Python systems, then promoted to AI/ML Engineer at Creative Bees in September 2026.

01Origin2 yrs

Robotics Engineer

Robomatic · Jun 2023 – Jul 2025

  • Engineered embedded systems & sensor control loops.
  • Conducted software QA, unit testing & testbench automation.
  • Instructed 150+ engineers in applied Python & robotics.
02Scale1 yr

Python Developer

Creative Bees · Aug 2025 – Aug 2026

  • Architected production Python backend microservices & REST APIs.
  • Built automated data pipelines with 99.9% reliability.
  • Prototyped initial ML integrations & OCR feature pipelines.
03CurrentPromoted

AI / ML Engineer

Creative Bees · Sept 2026 – Present

  • Leading autonomous Agentic AI loops (LangChain & LangGraph).
  • Building multimodal Vision-Language (VLM) document models.
  • Serving production deep learning models with sub-200ms latency.

Core competencies

04 disciplines

Generative AI & LLMs

RAG chat assistants with built-in lead capture, domain chatbots, and dense FAISS vector retrieval.

Video, Image & Vision AI

AI editing for YouTube Shorts with auto-captions, open-source diffusion image editing, and VLM document parsing.

Deep Learning Core

PyTorch & TensorFlow model training, custom loss architectures, and CUDA acceleration.

MLOps & Fast Inference

FastAPI microservices, Docker containerization, TensorRT optimization, and API orchestration.

Benchmark

Legacy workflow vs. autonomous AI

Deterministic / manual2 hrs / day

Fragile regex scripts & manual cross-app copy-pasting.

Neural / agentic pipeline1.2 sec

Nami Autonomous Agent cognitive loop plans & calls tools zero-shot.

6,000×faster, end to end
02 — Capabilities

The toolkit behind the models.

From multimodal vision-language models and deep neural networks to production-grade agentic systems — the frameworks I reach for every day.

01

Deep Learning & Core ML

Neural architecture design, mathematical optimization, and predictive modeling.

  • PyTorch
  • TensorFlow
  • Scikit-learn
  • CUDA
  • Keras
  • ONNX
  • NumPy
  • Pandas
02

Generative AI & Agentic Systems

Autonomous reasoning loops, multi-agent swarms, and retrieval-augmented systems.

  • LangChain
  • LangGraph
  • Agentic AI
  • RAG Pipelines
  • FAISS
  • Vector DBs
  • Gemini API
  • Qwen
  • Prompt Engineering
03

Computer Vision & Multimodal VLM

Visual perception, spatial segmentation, document understanding, and audio synthesis.

  • OpenCV
  • Vision-Language Models (VLM)
  • YOLOv8
  • FireRedOCR
  • CNNs
  • Face Anti-Spoofing
  • Stable Diffusion
  • Whisper AI
04

MLOps, Data Pipelines & Serving

Production packaging, low-latency model inference, and containerized deployment.

  • FastAPI
  • Docker
  • Hugging Face
  • Model Quantization
  • Streamlit
  • Git
  • Jupyter
  • Model Serving
  • Python
03 — Experience

Three roles, one direction.

Two years in robotics engineering, one year in production Python backend systems, and promoted to AI/ML Engineer in September 2026.

  1. Sept 2026 – PresentCurrent
    03

    AI / ML Engineer

    Creative BeesPromoted

    Coimbatore, Tamil Nadu, India

    • Promoted to lead production AI initiatives: architecting autonomous Agentic AI loops using LangChain and LangGraph for dynamic planning and multi-tool orchestration.
    • Developing high-throughput multimodal Vision-Language (VLM) pipelines (Qwen3, FireRedOCR) for unstructured document intelligence and zero-shot entity extraction.
    • Engineering and serving containerized Deep Learning inference microservices with FastAPI and Docker, achieving sub-200ms latency for production workloads.
    • Designing vector search retrieval architectures (RAG) powered by FAISS and dense transformer embeddings, boosting context precision and reducing model hallucination.
    • Agentic AI
    • Vision-Language Models (VLM)
    • PyTorch
    • LangChain
    • LangGraph
    • FastAPI
    • Docker
    • RAG Systems
    • FAISS
  2. Aug 2025 – Aug 20261 Year
    02

    Python Developer

    Creative Bees

    Coimbatore, Tamil Nadu, India

    • Built and scaled core Python backend microservices, RESTful APIs, and asynchronous message workers for business software applications.
    • Engineered automated data ingestion and preprocessing pipelines, ensuring 99.9% uptime and reliable integration across disparate data sources.
    • Integrated machine learning prototypes and OCR services into customer-facing web and mobile applications, laying the groundwork for full-scale AI adoption.
    • Conducted code reviews, unit testing, and Docker containerization to streamline CI/CD deployment cycles.
    • Python
    • FastAPI
    • Microservices
    • REST APIs
    • Docker
    • PostgreSQL
    • Data Pipelines
    • Backend Architecture
  3. Jun 2023 – Jul 20252 Years
    01

    Robotics Engineer

    Robomatic

    Coimbatore, Tamil Nadu, India

    • Served 2 years engineering embedded robotics control systems, sensor interfacing, and hardware-software communication protocols.
    • Trained and mentored 150+ students and technical professionals in Python fundamentals, software testing, quality assurance, and applied engineering.
    • Conducted extensive hands-on technical workshops covering machine learning concepts, computer vision fundamentals, and sensor automation pipelines.
    • Maintained lab hardware testbenches, automated test suites, and guided learners from theoretical algorithms to functional, working physical prototypes.
    • Robotics Systems
    • Embedded Systems
    • Sensors
    • Python
    • Software Testing
    • Quality Assurance
    • Technical Mentorship
04 — Selected work

Models & systems, shipped.

RAG assistants that turn chats into leads, an AI editor that turns raw footage into YouTube Shorts, open-source image editing, autonomous agents and vision models — each measured by what it does in production.

P/01RAG · Vector search · LLM · Lead capture
Chat → LeadAnswers questions, captures prospects

RAG Lead Assistant — AI chat that answers & captures leads

Retrieval-augmented chat assistant grounded in business knowledge. It answers visitor questions accurately from real documents and captures qualified leads — name, contact and intent — naturally inside the conversation.

  • RAG Systems
  • LLM
  • Vector DB
  • Lead Generation
  • Chatbot
P/02Speech AI · Multimodal understanding · Auto-render
Raw → ShortCuts, captions & highlights — automated

AI Shorts Editor — Raw footage to YouTube Shorts

Upload raw clips and a script; the AI understands the video and audio, transcribes the speech, chooses the cuts, and renders vertical YouTube Shorts with animated captions and highlighted keywords.

  • Video AI
  • Speech-to-Text
  • Auto Captions
  • Multimodal AI
  • YouTube Shorts
P/03Diffusion · Segmentation · Inpainting
Open-sourceSelf-hosted generative image editing

AI Image Editor — Prompt-driven editing on open-source models

Carvona — an image editing pipeline built entirely on open-source diffusion and segmentation models. Select, remove, replace and restyle parts of an image with plain-language prompts, including professional automotive photo editing.

  • Stable Diffusion
  • Inpainting
  • Segmentation
  • Open-Source Models
P/04LangGraph · LangChain · Cognitive loops
1.2sMulti-step planning

Nami Agent — Autonomous multi-agent loop

Autonomous workflow engine with dynamic tool-calling, calendar orchestration, and self-correcting agent execution loops.

  • Agentic AI
  • LangGraph
  • LangChain
  • Python
P/05Qwen3 · FireRedOCR · Vision-Language
280msStructured JSON extraction

Buzz Card — Multimodal VLM document parser

Vision-Language pipeline extracting business cards and unstructured documents into CRM entities zero-shot.

  • VLM
  • Qwen3
  • FireRedOCR
  • Computer Vision
P/06Deep CNN · OpenCV · Biometrics
99.2%Anti-spoof accuracy

Face Anti-Spoofing — Real-time liveness detection

Production biometric security model distinguishing live human faces from 2D photos, masks, and screen replays in real time.

  • TensorFlow
  • CNN
  • OpenCV
  • Biometrics
P/07Stable Diffusion · ControlNet · YOLOv8
3.5sPhotorealistic inpainting

Diffusion Vehicle Studio — Segmentation-guided inpainting

Generative AI pipeline combining YOLOv8 segmentation with ControlNet to generate studio-grade vehicle visual presentations.

  • Stable Diffusion
  • ControlNet
  • YOLOv8
  • Inpainting
P/08FAISS · Dense embeddings · Gemini 1.5
CitedGrounded citation retrieval

Legal AI & RAG Engine — Grounded legal intelligence

Legal intelligence platform indexing Constitution & IPC legal documents with dense vector search for zero-hallucination answers.

  • RAG Systems
  • FAISS
  • LangChain
  • Vector DB
P/09Neural vocoder · Speaker conditioning
Zero-shotVoice prosody transfer

Neural Voice Cloning — Speech synthesis

Deep audio synthesis pipeline implementing speaker embeddings and neural vocoders for natural conversational speech.

  • Audio ML
  • TTS
  • Neural Voice
  • PyTorch
05 — Open source

AI repositories

github.com/prasannab4362
Contribution streakPrasanna B GitHub contribution streak statistics
06 — Teaching & STEM

Every Sunday, robotics & AI.

My passion beyond production AI is STEM education. Every Sunday I teach robotics and AI to students in my home area in Coimbatore — from Class 2 children building their first robot to older students training their first AI model. Across classes and workshops I have mentored 150+ students and professionals.

SundayEvery week · Coimbatore

Sunday Robotics & AI Classes for young learners

Hands-on sessions in my neighbourhood where students build and program robots, explore sensors and electronics, and learn how AI sees, listens and decides — explained simply, starting from Class 2.

  • Robotics
  • Electronics
  • Coding
  • Artificial Intelligence
  • Class 2 & up
Enquire on WhatsApp
  1. 01Class 2 – 5

    Curiosity

    First robots, simple machines and block-based coding — making logic and problem-solving feel like play.

    • Simple robots
    • Block coding
    • How machines think
  2. 02Class 6 – 10

    Building

    Arduino, sensors and motors, then a first real programming language — learning by making robots move.

    • Arduino
    • Sensors & motors
    • Python basics
  3. 03Class 11 – College

    Engineering

    Python, Raspberry Pi and computer vision projects that connect classroom maths to working AI.

    • Python
    • Raspberry Pi
    • Computer vision
  4. 04Graduates & professionals

    Applied AI

    Machine learning, software testing and production AI workflows — the skills used on real engineering teams.

    • Machine learning
    • Software testing
    • AI engineering
150+Students & professionals mentored
Build first
Every class ends with something that works — a circuit, a robot, a model.
Explain simply
If a Class 2 student can't follow it, the explanation isn't finished yet.
Keep learning
I teach what I'm learning; Sunday classes and production AI sharpen each other.
07 — Contact

Let's build something intelligent.

Chat on WhatsAppProjects, roles, collaborations or Sunday classes — tap to message me.

Have a project, job opening, or collaboration in mind? Send a message and let's turn agentic AI concepts into production realities.