Senior Applied AI Engineer · Bengaluru

I build production AI systems from retrieval to runtime.

Seven years across GenAI, LLM systems, computer vision, and distributed data platforms. I turn complex, unstructured data into reliable products that perform at scale.

View résumé Connect on LinkedIn GitHub ↗

Currently

Senior AI Engineer, Data Platforms

TGS (Steerwise) · Remote, India · Jul 2022 to Present

production-ai.systems

$
7+ Years in applied AI
Petabyte Scale data engineering
Cloud ↔ Edge End-to-end deployment

01 · Profile

Applied AI, all the way to production.

I design and scale production GenAI and LLM ecosystems, with a focus on multimodal RAG, structured extraction, retrieval quality, and trustworthy evaluation.

My foundation is in C++ and computer vision, including segmentation, detection, tracking, and foundation models. I extend that work through edge inference, distributed systems, and cloud platforms. That breadth lets me solve the whole problem: data, model, evaluation, infrastructure, and runtime performance.

01

Build for evidence. Evaluation, observability, and benchmarking belong inside the architecture.

02

Engineer the full path. From raw, high-entropy data to a reliable production interface.

03

Optimize where it matters. Retrieval quality, latency, throughput, memory, and cost.

02 · Expertise

Systems that reason, retrieve, and scale.

Data Retrieve Reason Evaluate
RAG Multimodal Vector Search LLMs

Production GenAI & RAG

End-to-end retrieval systems over complex enterprise data: chunking and embeddings, IVF/FAISS vector search, tool orchestration, structured extraction, prompt engineering, and transformer-based workflows.

From context engineering to deployment
RAGASOTLPGrafana

AI Evaluation & Observability

Faithfulness, retrieval precision, hallucination monitoring, drift analysis, and operational telemetry for latency, throughput, and cost.

DaskGeoPandasGCPAWS

Data Platforms & Distributed Systems

Petabyte-scale ingestion and processing, workflow orchestration, cloud-native storage, data quality, Kubernetes, and event-driven systems.

TensorRTONNXJetsonC++

Computer Vision & Edge AI

Detection, tracking, segmentation, camera systems, and low-latency deployment through quantization, pruning, profiling, and hardware-aware optimization.

MLflowAirflowKubeflowDocker

MLOps & Performance Engineering

Reproducible pipelines, CI/CD, model lifecycle tooling, automated retraining, and root-cause analysis across cloud and edge workloads.

03 · Selected systems

Complex problems, built into working systems.

01 / Multimodal AI

Seismic AI assistant

Integrated ViT/MAE computer vision with LLMs to enable semantic search, fault segmentation, and intelligent metadata extraction across petabyte-scale seismic datasets.

Vision + language + high-volume domain data

02 / Trustworthy AI

Evaluation & observability framework

Built systematic LLM evaluation with RAGAS and operational monitoring with OTLP/Grafana to track faithfulness, retrieval quality, hallucinations, latency, throughput, and cost.

Quality signals made operational

03 / Edge intelligence

Real-time vision on constrained hardware

Deployed driver monitoring, video analytics, and tracking pipelines across NXP and Samsung automotive SOCs and NVIDIA Jetson using TensorRT, ONNX, DeepStream, and INT8/FP16 optimization.

Low-latency inference from car cabin to CCTV

04 · Experience

Seven years of production engineering.

  1. Current

    Senior AI Engineer, Data Platforms

    TGS (Steerwise) · Remote, India

    Via Affine Analytics, Jul 2022 to Jun 2024; direct with Steerwise, Jul 2024 to Present

    • Architected multimodal AI and production RAG systems for semantic search, fault segmentation, structured extraction, and natural-language access to seismic metadata.
    • Built an AI evaluation and observability framework using RAGAS, OTLP, and Grafana.
    • Scaled SEG-Y ingestion and distributed processing across petabyte-scale datasets with Dask, NumPy, and GeoPandas.
    • Operated cloud-native data platforms across GCP and AWS with Kubernetes, Airflow, serverless execution, event-driven messaging, and CI/CD.

  2. Software Engineer, Machine Learning

    Visteon Corporation · Bengaluru

    • Led driver-monitoring development for automotive ADAS, including RetinaFace tuning and in-cabin inference.
    • Ported and optimized models for NXP and Samsung SOCs using pruning, layer fusion, and INT8/FP16 quantization.
    • Built 360° surround-view and photometric-alignment systems in C++ with OpenCV, Eigen, and OpenGL.

  3. Machine Learning Engineer, Computer Vision

    Beyond Drops · Bengaluru

    Built fingerprint biometrics and browser-based eKYC systems using deep learning, GCP, Docker, OpenPose, MediaPipe, and TensorFlow.js.

  4. Machine Learning Engineer, Computer Vision

    Integration Wizards · Bengaluru

    Delivered real-time CCTV intelligence and custom multi-object tracking on NVIDIA Jetson using DeepStream, GStreamer, TensorRT, ONNX, CUDA, and optimized C++.

  5. Software Engineer, Machine Learning Intern

    ApplicateAI · Bengaluru

    Built conversational AI assistants with Rasa and spaCy, and developed demand-forecasting models using deep regression networks.

05 · Toolkit

Deep across the AI stack.

Python C++ PyTorch TensorFlow RAG / FAISS / IVF RAGAS LangChain / LangGraph ViT / MAE OpenCV TensorRT / ONNX NVIDIA DeepStream Dask / GeoPandas Kubernetes Docker Airflow / Kubeflow MLflow GCP / AWS OTLP / Grafana

06 · Contact

Let’s build AI that holds up in production.

I’m interested in ambitious applied AI work across GenAI, retrieval, evaluation, computer vision, edge inference, and large-scale data platforms.

Email me LinkedIn GitHub Résumé