Computer Vision / Machine Learning

GEMBA CONCEPTS

Experience: ~3–5 years Type: Full-time

AI/ML Engineer

Location: Bengaluru, India (Hybrid)

About Gemba Concepts

Gemba Concepts is a lean manufacturing and technology consulting firm helping clients across pharma, manufacturing, and logistics

modernize how they operate. We build production systems that sit close to the shop floor — warehouse management, manufacturing

traceability, and an applied AI/ML platform whose flagship use cases are visual quality inspection and predictive maintenance. We’re a

tight engineering team that ships real systems for demanding, often regulated, environments.

The Role

We’re looking for an AI/ML Engineer to take ML capabilities from prototype to production. You’ll own models end-to-end — framing the

problem with stakeholders, building and validating the model, and deploying it as a reliable service that holds up against real-world, messy

industrial data. This is a hands-on building role, not a pure research seat: your work goes into client-facing systems.

What You’ll Do

Build and ship computer vision models for visual quality inspection (defect detection, classification, segmentation) that perform under

real factory lighting, throughput, and edge-case conditions.

Develop predictive maintenance models using sensor/time-series data — anomaly detection, remaining-useful-life estimation, failure

prediction.

Own the full ML lifecycle: data pipelines, feature engineering, training, evaluation, and deployment, with proper versioning and monitoring.

Deploy and serve models in production on Azure (AKS), and keep them healthy — track drift, retraining triggers, and latency.

Integrate LLM-based capabilities (we use the Claude API and self-hosted open models) into delivery and product workflows where they

add leverage.

Collaborate with product, engineering, and domain experts to translate fuzzy operational problems into well-scoped ML solutions — and to

know when ML is not the right answer.

Communicate results and limitations clearly to non-ML stakeholders, including clients.

What We’re Looking For

3–5 years of hands-on experience building and deploying ML models in production (not just notebooks or coursework).

Strong Python and the modern ML stack — PyTorch or TensorFlow, scikit-learn, NumPy/Pandas.

Solid grounding in at least one of: computer vision (CNNs, object detection/segmentation, image preprocessing) or time-series /

anomaly detection.

Practical MLOps experience: containerization (Docker), model serving, experiment tracking, and deploying on a cloud platform — Azure /

Kubernetes (AKS) is a strong plus.

Comfort working with imperfect, real-world data — labeling strategy, class imbalance, data drift, and validation that reflects production

reality.

Good engineering hygiene (Git, testing, code review) and the ability to write code others can build on.

Nice to Have

Experience with industrial / manufacturing data or regulated environments (pharma, 21 CFR Part 11 awareness).

Hands-on LLM integration experience — RAG, prompt engineering, working with APIs or self-hosted models (vLLM, Qwen, etc.).

Edge deployment experience (running CV models on-device / near the line).

Exposure to data pipeline tooling and orchestration.

What You’ll Get

Real ownership of ML systems that go into production for serious clients.

A lean, senior-heavy team where you ship fast and learn across the stack.

Direct exposure to applied AI in manufacturing — a domain where the work has tangible, physical impact

AuxoAI is hiring a Senior Applied AI Engineer to design and deploy production-grade computer vision systems that operate reliably in real-world environments.

This role focuses on building end-to-end visual intelligence systems, combining deep learning, classical computer vision techniques, and multimodal models. It is not limited to model training and requires strong ownership of system design, deployment, and real-world performance.

You will work on systems that perform perception, understanding, and reasoning over visual data, and integrate these capabilities into larger AI platforms and agent-based workflows.

You will also work on problems where existing approaches may not be sufficient, and will be expected to combine deep learning, geometric methods, and multimodal reasoning to build robust, production-grade systems.

Location – Mumbai / Bangalore / Hyderabad / Gurgaon (Hybrid – 3 days per week in office)

Responsibilities:

• Design and deploy computer vision systems for tasks such as:
• Object detection, segmentation, and tracking
• Scene understanding and structured perception
• Video understanding and temporal reasoning
• Build and optimize models using architectures such as:
• CNNs (ResNet, EfficientNet)
• Vision Transformers (ViT, Swin, DeiT)
• Detection/segmentation models (YOLO, DETR, Mask R-CNN)
• Develop multimodal systems combining vision and language:
• CLIP-style models
• Vision-language models (VLMs)
• Visual grounding and captioning systems
• Implement algorithms for:
• Multi-object tracking (SORT, DeepSORT, ByteTrack)
• Feature matching and representation learning
• Temporal modeling (RNNs, Transformers for video)
• Apply geometric and classical computer vision methods where relevant:
• Camera calibration
• Epipolar geometry
• Pose estimation
• 3D reconstruction or depth estimation
• Optimize systems for:
• Low-latency, real-time inference
• Throughput and scalability
• Edge and distributed deployment
• Design and build data pipelines for:
• Annotation workflows
• Dataset curation
• Synthetic data generation
• Integrate vision systems into:
• Multimodal AI pipelines
• Agent-based systems
• Decision-making workflows

Requirements:

• 5+ years of experience building computer vision systems in production environments
• Strong experience with deep learning frameworks (PyTorch / TensorFlow)
• Hands-on experience with:
• Detection, segmentation, or tracking systems
• Model training, fine-tuning, and evaluation
• Strong understanding of:
• Representation learning
• Loss functions (contrastive loss, focal loss, etc.)
• Evaluation metrics (mAP, IoU, precision/recall)
• Experience building and deploying end-to-end vision systems, not just training models

Candidates whose primary experience is limited to academic projects or model experimentation without real-world deployment may not be a fit for this role.

Nice to Have:

• Experience with multimodal systems (vision + language)
• Familiarity with models such as:
• CLIP, BLIP, Flamingo, or similar
• Experience with 3D vision:
• NeRFs
• SLAM
• Point clouds
• Experience with video understanding:
• Action recognition
• Event detection
• Experience building data engines:
• Active learning
• Hard negative mining
• Experience working with large-scale datasets and distributed training pipelines

Hi,

Greetings !!

We/re are looking for someone who has Hands-on experience with CV/ML

The location for the same is Bangalore.

Requirements

• 11–14 years total experience
• Computer Vision – strong hands-on experience
• Object Detection – YOLO(Preferred), Faster R-CNN, SSD, etc.
• Image Processing – OpenCV, image enhancement, segmentation, feature extraction
• Machine Learning / Deep Learning – CNNs, model training, evaluation, optimization
• AI/ML – production-level AI solution development
• LLM / GenAI – practical exposure to LLMs, multimodal AI, RAG, VLMs, or GenAI
• Python – strong programming skills
• Model deployment – preferably TensorRT, ONNX, Docker, Kubernetes, cloud, or edge deployment
• Bangalore – candidate should be based in / willing to work from Bangalore

Preferred

• Vision Transformers / ViT
• YOLOv8/YOLOv9/YOLOv10/YOLO11
• PyTorch / TensorFlow
• NLP / LLM / VLM
• Generative AI
• CUDA / GPU optimization
• Edge AI / NVIDIA
• Experience leading CV/AI projects or teams

If interested, Share CV at: snigdhaattheratebeanhr.com

About the role

We give you a product design and a brief. You turn it into backend software that analyses images and user input and hands our frontend team the structured data it needs.

Most of the code gets written by AI agents like Claude Code and Codex, with you directing them. That makes you responsible for every line they produce, including the wrong ones. Your value is knowing what to ask for and catching what the agent got wrong. If “the AI did it” sounds like a reasonable defence to you, this job will frustrate you.

What you will do

1. Read a design and work out what data each part of it needs.
2. Build the backend logic that produces that data, with AI coding agents doing most of the typing.
3. Set up open-source computer vision and AI models, run them on real images, and decide whether they are good enough.
4. Check your own output before anyone reviews it. Wrong values, invented APIs and messy code should stop with you.
5. Ship small, well-typed pull requests that get through code review.

You won’t have to come up with the product itself. We do that part, and your job is to make it work and keep it maintainable.

You will fit if you

1. Write clear, precise English in prompts, pull requests and messages. This is the first thing we screen for.
2. Use Claude Code, Codex or Cursor every day, and can show us how you break a task down for an agent.
3. Know enough computer vision, machine learning and data engineering to ask for the right thing and to recognise a wrong answer when you see one.
4. Are a solid Python engineer who understands typing and async, and who spots bad patterns in a diff quickly.
5. Look at the actual output every time. Code that runs proves very little.

Nice to have

1. Experience with OpenCV or similar image libraries.
2. Experience with FastAPI, workflow engines, or cloud storage.
3. Experience getting structured outputs from LLM APIs.
4. A side project where you got an open-source model running and evaluated it.

How to apply

Send your CV and answers to these 3 questions, 150 words maximum each:

1. Describe one time an AI coding agent gave you code that looked right but was wrong. How did you find out?
2. You must estimate the real-world size of an object from a single photo. What steps and checks would you ask an AI agent to build?
3. Paste one real prompt you recently gave an AI coding agent, and tell us what happened next.

Include a link to one piece of your work: a repository, pull request, write-up or demo.

About Naicos

Naicos, a fast-paced startup, builds AI-native products for algorithmic commerce: the future of how e-commerce runs. Our first products are already live with paying customers, and we are shipping new ones continuously.

Your Role

You will drive the research behind our imaging products, finding approaches to hard, unsolved problems in product and apparel imagery that work at production scale. You will run the experiments, prove what is viable, and hand a working approach to the engineering team.

Who We Are Looking For

•     Total experience: 3 years or more, with a strong research orientation

•     Deep learning frameworks in Python: PyTorch or TensorFlow

•     Image processing in Python: OpenCV, Pillow, scikit-image

•     Working knowledge of diffusion and other image generation models

We are looking for a strong research or research-student profile: someone who investigates, experiments and proves an approach, working closely with the AI Architect. Someone who is driven to build solutions, not just desk research.

AI Skills and Experience

•     Computer vision: classical CV alongside deep learning.

•     Segmentation, image-to-image translation, geometry and lighting; 

•     Generative imaging: diffusion models, conditioning and control, fine-tuning and LoRA,

•     Reads academic papers, judges what is reproducible, and turns one into a working prototype in days

Good to have

•     3D and rendering; published research or open-source contributions; model optimisation for inference cost

Research and innovative problem solving

•     Comfortable where there is no known answer, and defines the approach yourself

•     Solves problems inventively rather than reaching for the biggest model; many results come from classical image processing, fitment and geometric transformation

Other Relevant Skills and Experience

•     Designs experiments: baselines, measurable success criteria, honest reporting of negative results

•     Explains findings to a non-research audience and guides engineers to production

•     Git and reproducible experiment tracking (Weights & Biases, MLflow or similar)

Educational Qualification

•     BE / B.Tech / ME / M.Tech in Computer Science

•     BE / B.Tech / ME / M.Tech in any discipline with proven Computer Vision coursework or work

•     MSc / MS in Computer Science, Maths, Statistics or Computer Vision

•     PhD in Computer Vision or Machine Learning: an advantage, not a requirement

•     Reputed Tier 1 university preferred

• Senior Machine Learning Engineer – Vertex AI
• Senior Generative AI Engineer – GCP
• Google Cloud AI/ML Engineer
• MLOps Engineer – Vertex AI
• LLM Engineer – RAG, Gemini, and Vertex AI

About the role

We are building AI systems that read, understand and act on real business documents, bank statements, financial reports, policy documents and forms and putting them into production where accuracy and cost both matters.

This is not a research role and it is not a prompt-writing role. You will own features end to end: pick and deploy open-source models, build the pipelines around them, measure whether they actually work on our documents, drive the cost per document down, and keep the whole thing running in production.

You will work closely with the engineering and product teams, and your work will be directly used by business users from day one.

What you will do

Deploy and evaluate open-source models

• Select, deploy and benchmark open-source LLMs and vision-language models for specific, narrow use cases not general chat.
• Build evaluation sets from real documents and define what "good" means numerically (field-level accuracy, extraction recall, hallucination rate) before shipping.
• Run structured comparisons between models and approaches, and write up the trade-offs so the team can make a decision.
• Apply quantization, batching and other optimizations to fit models into a sensible GPU budget.

Build and optimize AI orchestration

• Design multi-step pipelines that combine deterministic code, ML models and LLM calls and know when not to use an LLM.
• Optimize for latency, cost and reliability: caching, batching, request routing, fallback tiers, retries and graceful degradation.
• Instrument pipelines so failures are visible and traceable rather than silent.

Ship to production

• Package models and services with Docker, expose them behind clean APIs, and deploy them to our GPU and CPU infrastructure.
• Handle the unglamorous production concerns: cold starts, timeouts, concurrency limits, versioning, rollback and monitoring.
• Own on-call-style responsibility for the AI features you build, including cost tracking.

Must-have skills

Programming & engineering

• Strong Python: type hints, async/await, dataclasses/Pydantic, clean module design, testing.
• REST API development with FastAPI (or Flask/Django with a willingness to move to FastAPI).
• Git, code review discipline, and the ability to write code someone else can maintain.
• Comfortable in Linux and on the command line.

Machine learning fundamentals

• Working knowledge of PyTorch and the Hugging Face ecosystem (transformers, tokenizers, accelerate).
• Understanding of inference-time concepts: tokenization, context windows, batching, precision (FP16/BF16/INT8), memory footprint.
• Ability to read a model card and a paper well enough to judge whether a model fits a use case.

Document processing

• Hands-on experience with at least two of: pypdfium2, PyMuPDF, pdfplumber, pdfminer.six, Docling, Unstructured, Surya, DocTR, LayoutLM family.
• Practical OCR experience (Tesseract, PaddleOCR, or a cloud OCR) and an understanding of when OCR is the wrong tool.
• Experience extracting tables from PDFs and dealing with merged cells, multi-line rows, and inconsistent column layouts.

Strongly preferred

You will be a much stronger candidate with any of these. We do not expect all of them.

Model serving & optimization

• vLLM, TGI, Ollama, llama.cpp, or Triton Inference Server.
• Quantization formats and tooling: GGUF, AWQ, GPTQ, bitsandbytes, ONNX Runtime, INT8 export.
• Serverless GPU platforms: Modal, RunPod, Replicate, Baseten including cold-start and container-lifecycle management.
• LoRA / QLoRA fine-tuning with PEFT for narrow, task-specific improvements.

Vision-language models

• Practical use of open VLMs: Qwen2.5-VL, InternVL, Granite Vision, Molmo, Phi-Vision, or similar.
• Awareness of where VLMs hallucinate especially on numeric and financial content and patterns for constraining them (using the model for layout only, sourcing values from the text layer, constrained decoding).

Orchestration & pipelines

• Workflow orchestration: Dagster, Airflow, Prefect, or Temporal.
• Async job patterns: Celery, RQ, or platform-native spawn/poll patterns.
• LLM orchestration frameworks (LangGraph, LlamaIndex, Haystack) with the judgement to know when plain Python is a better answer.
• Structured output enforcement: Instructor, Outlines, XGrammar, JSON schema / tool-use modes.

Evaluation & observability

• Building golden datasets and regression suites for extraction tasks.
• Eval tooling: promptfoo, DeepEval, Ragas, or in-house harnesses.
• LLM tracing and monitoring: Langfuse, Arize Phoenix, LangSmith, OpenTelemetry.

Nice extras

• Rule engines and policy evaluation (Open Policy Agent / Rego, Drools, rule-engine).
• Experience in fintech, lending, insurance or accounting documents.
• Handling of PII and data-security practices in document pipelines.
• Contributions to open-source ML or document-processing projects.

Why join us

• Real production ownership from month one your work goes to actual users, not a demo.
• Genuinely hard technical problems in document AI, not wrappers over an API.
• Small team, short decision cycles, direct access to leadership.
• Budget and freedom to evaluate and adopt new open-source models as they land.

To apply: send your CV along with a short note on one AI system you have taken to production what it did, what the accuracy was, and what broke.

About the Role We are seeking a highly technical, hands-on Senior AI/ML Tech Lead to drive the design, development, and deployment of cutting-edge Generative AI applications. In this dual-impact role, you wi l act as a primary individual contributor architecting core AI engines while simultaneously leading a team of engineers through task alocation, code reviews, and technical mentorship. The ideal candidate bridges the gap between state-of-the-art AI research (LLMs, Agentic frameworks, Advanced RAG, OCR) and production-grade ful-stack engineering (Python, FastAPI, React).

Key Responsibilities

Technical Leadership & Team Management (40%)

● Technical Oversight: Lead a team of AI, backend, and ful-stack engineers; alocate tasks, establish sprint priorities, and ensure timely delivery.

● Code Quality & Reviews: Conduct rigorous code reviews to maintain high engineering standards, security, performance, and scalability across AI and fu l-stack codebases.

● Architecture & Governance: Design end-to-end system architectures for AI solutions, ensuring seamless integration between frontend interfaces, backend APIs, and AI models.

● Mentorship: Guide and upskil team members on modern software practices, LLM engineering, and agentic design patterns. Hands-On Engineering & Development (60%)

● Generative AI & Agentic Systems: Architect, build, and optimize LLM-powered applications, multi-agent workflows (e.g., CrewAI, AutoGen, LangGraph), and autonomous AI agents.

● RAG & OCR Pipelines: Design and deploy advanced RAG (Retrieval-Augmented Generation) architectures and document processing pipelines utilizing OCR techniques (e.g., LayoutLM, PaddleOCR, Tesseract, Vision LLMs) to extract structured data from unstructured sources.

● Backend Systems: Build robust, asynchronous, high-throughput microservices and RESTful APIs using Python and FastAPI.

● Frontend Integration: Colaborate on or build modern web interfaces using React (e.g., Control Towers, operations dashboards, interactive chat interfaces).

● MLOps & Vector DBs: Oversee model deployment, prompt engineering, fine-tuning, vector database integration (Pinecone, Qdrant, Chroma, PGVector), and cloud infrastructure setup (Azure/AWS).

Required Qualifications & Skills

● Overall Experience: 8 to 10 years of professional software engineering experience.

● AI/ML Domain Experience: 3 to 4+ years of dedicated, hands-on experience building and deploying AI/ML, OCR, and Generative AI solutions in production.

● Core Technical Stack: ○ Generative AI & LLMs: Extensive experience with commercial and open-source LLMs (OpenAI, Anthropic Claude, Llama), Agentic frameworks (LangChain, LlamaIndex, AutoGen, CrewAI), and LLM evaluation frameworks (LangSmith, TruLens, Ragas). ○ RAG & Unstructured Data: Strong knowledge of hybrid search, re-ranking, chunking strategies, vector databases, and document inte ligence workflows. ○ OCR & Vision Techniques: Hands-on experience with OCR engines (Tesseract, PaddleOCR, Azure Document Inteligence) and Multi-Modal/Vision LLMs for document extraction. ○ Backend: Deep expertise in Python and asynchronous frameworks (FastAPI, AsyncIO). ○ Frontend: Working proficiency in React (TypeScript/JavaScript) for building interactive web UI components. ○ Cloud & DevOps: Hands-on experience with cloud platforms (Azure / AWS), Docker, Kubernetes, and CI/CD pipelines.

Preferred / Good-to-Have Skills

● Experience with cloud-native data platforms (e.g., Microsoft Fabric, Snowflake, Azure SQL).

● Familiarity with cost optimization and latency reduction techniques for LLM inference (caching, semantic routing, model quantization).

● Prior experience in client-facing technical leadership or agile consulting environments.

What We Offer

● Opportunity to lead and build high-impact, state-of-the-art Generative AI systems.

● Colaborative engineering culture with room for technical ownership and direct business impact.

● Flexible work arrangements and competitive compensation package.

Job Description:

We are seeking a versatile and highly skilled Lead AI/ML Engineer with deep expertise in Generative AI (GenAI) and Large Language Models (LLMs). This role requires a leader who can take full ownership of the

AI lifecycle—from initial architectural design to final production execution. You will lead the development of scalable AI-powered applications, demonstrating exceptional execution skills and the ability to deliver high-performance results under pressure in demanding production environments.

Machine Learning & LLM Capability:

 End-to-End ML Engineering: Build and manage comprehensive ML pipelines, including data ingestion, preprocessing, training, and evaluation using frameworks like PyTorch, TensorFlow, and Scikit-learn.  Advanced LLM Systems: Design and implement sophisticated LLM-based applications such as autonomous agents, chatbots, and complex automation tools.

 Generative AI Specialization: Architect and optimize Retrieval-Augmented Generation (RAG) pipelines using vector databases like FAISS, Pinecone, or Weaviate.

 Model Optimization: Fine-tune open-source and proprietary models (e.g., LLaMA, GPT) using advanced techniques like LoRA, QLoRA, or instruction tuning.

 Agentic Frameworks: Develop complex agentic workflows utilizing frameworks such as LangChain or LlamaIndex.

 Prompt Engineering: Implement expert-level prompt engineering, tool/function calling, and structured output generation.

 Project Ownership & Execution

 Full Lifecycle Ownership: Take complete accountability for the full ML and GenAI lifecycle, spanning data processing, model development, monitoring, and optimization.

 Architectural Leadership: Drive strategic architectural decisions for AI platforms, ensuring they are modular, scalable, and maintainable.

 Execution Excellence: Write clean, high-performance Python code following strict OOP principles and manage CI/CD pipelines for seamless project execution.

 Leadership & Mentoring: Act as a key technical leader, managing stakeholders and mentoring team members to ensure all project milestones are met with quality.

 System Integrity: Manage model and prompt versioning, experiment tracking, and comprehensive documentation for all pipelines and workflows.

 Performance Under Pressure

 Production Reliability: Ensure all AI systems maintain extreme scalability and performance under heavy production workloads, including both batch and real-time processing.

 High-Pressure Optimization: Rapidly optimize inference latency and system costs for ML and LLM systems to meet urgent business and technical requirements.

 Proactive Problem Solving: Apply strong analytical thinking to address complex challenges such as system drift, hallucinations, and latency in fast-paced environments.

 Robust Guardrails: Implement and manage strict evaluation frameworks and feedback loops to maintain system quality under stress.

Qualifications:

 Bachelor’s or Master’s degree in Computer Science, AI, ML, or a related field.

 Proven expertise in Python, system design, and scalable AI/ML architecture.

 Deep knowledge of NLP, Computer Vision, and Deep Learning models.

 Hands-on experience with Docker, Kubernetes, MLOps, and major cloud platforms (AWS, GCP, or Azure).

Position: Computer Vision Engineer

Experience: 2–3 Years

Location: Bengaluru, Karnataka

Employment Type: Full-time

About the Role

We are seeking a highly motivated Computer Vision Engineer to join our autonomy and avionics team. The role involves developing, implementing, and validating computer vision models and algorithms and pipelines for UAVs operating in both GNSS-available and GNSS-denied environments.

The ideal candidate should have a strong foundation in theory of deep learning and machine learning, strong understanding of electromagnetic spectrum, imaging fundamentals, camera principles, and mathematical concepts with hands-on experience in implementing these algorithms on embedded or real-time systems.

Key Responsibilities

• Design, develop, and optimise AI Models
• Make custom CNNs/ modify existing CNNs to suit specific problems at hand
• Handle end-to-end training flow
• Implement end to end inference pipelines on standard PCs as well as on embedded systems
• Understand performance benchmarks and assess the accuracy and inference times
• Implement traditional image processing algorithms
• Factor the code to leverage underlying hardware architecture
• Prune the networks for efficiency
• Integrate the system within the application framework using C++
• Work closely with perception, controls, embedded software, and systems engineering teams.

Required Qualifications

• B.E./B.Tech/M.E./M.Tech in Computer Science and Engineering, Electronics, ECE, Mechatronics, or a related discipline.
• 2–3 years of experience in relevant area
• Strong understanding of: Linear Algebra, Probability and Statistics, AI-ML-DL fundamentals, Image processing, Camera Functioning
• Strong programming skills in C++ and Python.
• Experience with MATLAB for algorithm development and validation.
• Familiarity with Linux development environments.
• Experience with Git version control.

Preferred Skills

• Experience with Camera, IMU Calibration and Synchronisation
• Experience with multi-sensor fusion.
• Experience working with NVIDIA devices
• Experience on FPGA will be an added advantage
• Full understanding of Git functionality
• Exposure to airborne software development processes and coding standards (e.g., MISRA C++).

Personal Attributes

• Strong analytical and problem-solving skills.
• Ability to work independently on challenging technical problems.
• Good communication and documentation skills.
• Passion for solving challenging problems
• Willingness to participate in field trials and flight testing.
• Team player