Setup Local AI Environment: Conda, GPU, PyTorch

You spent two days installing PyTorch, and it still sees CPU? A familiar situation. A proper local environment is the foundation of an AI engineer's productivity. Over 5 years, we have set up more than 50 AI environments and developed a clear recipe that cuts setup time from two days to one hour. Dependency management, CUDA versions, and environment isolation are the basics. Without them, experiments are irreproducible and p99 latency suffers. We use proven tools and patterns so you can focus on models, not the environment.

Why Environment Isolation Is Critical for AI Projects

Each ML project requires its own stack: some use PyTorch 2.0, others TensorFlow 2.15. Conflicts between CUDA, cuDNN, and Python versions are a common time sink. We use Conda as the primary environment manager. Unlike venv, Conda manages not only Python packages but also system dependencies (CUDA, NCCL). This allows fully isolated environments:

# Example: create environment for an NLP project
conda create -n nlp python=3.11
conda activate nlp
pip install torch torchvision --index-url https://download.pytorch.org/whl/cu121
pip install transformers datasets peft accelerate

This approach guarantees that you can work on a PyTorch classification project and a TensorFlow generative project in parallel on the same machine without breaking dependencies.

How to Speed Up Development with Remote GPUs

For resource-intensive tasks (fine-tuning LLMs, training from scratch), local GPUs are often insufficient. We set up a hybrid scheme: code stored locally (Git), data in S3/GCS, compute on cloud instances (SageMaker, Vertex AI). Connection via VS Code Remote SSH lets you edit code locally and run it on a powerful GPU machine. This is cheaper than keeping a local A100 and more convenient than working in a browser-based Jupyter.

What's Included in AI Development Environment Setup

We offer a turnkey service: from analyzing your current environment to full configuration. As a result, you get:

• Isolated Conda environments per project with pinned versions (Python, CUDA, libraries)
• IDE configuration (VS Code, PyCharm) with linters (black, isort, mypy) and debugger
• Jupyter Lab integration with remote kernel capability
• DVC configuration for data versioning and MLflow for experiment tracking
• GPU profiling scripts (PyTorch Profiler) and bottleneck analysis
• Remote GPU access setup (SSH config, SSHFS)
• Network infrastructure documentation (ports, proxies, VPN)
• Team training: we show how to work with the new environment

Common Mistakes in Self-Setup

• Installing PyTorch for CPU instead of CUDA — you download the version without GPU support. Check torch.cuda.is_available().
• Mixing pip and conda — conda does not track pip installations. Use only pip inside a conda environment.
• Missing .gitignore for datasets and models — cluttering the repository. DVC solves this.
• Pinning only Python dependencies — forgetting system libraries (CUDA, cuDNN). Conda's environment.yml includes everything.

How We Set Up Your Environment in 6 Steps

1. Audit current environment — identify conflicts, unused dependencies, suboptimal GPU settings.
2. Design architecture — choose stack (Conda vs Docker, DVC vs Git LFS, MLflow vs W&B).
3. Build base image — create environment.yml with full dependency list, including system packages.
4. Integrate tools — VS Code, Jupyter, DVC, MLflow, pre-commit hooks.
5. Test reproducibility — verify that the environment deploys from scratch in 15 minutes.
6. Deliver documentation — hand over a README with instructions and configuration files.

Guarantees and Experience

We have worked with AI infrastructure for over 5 years. Our engineers are AWS ML Specialty certified and have set up environments for projects with up to 10^8 parameters. We guarantee that after setup you won't face import errors or CUDA incompatibility. If an issue arises, we fix it within 24 hours. We assess your project in one day: just write to us with a description of your current stack. Order the setup today and save hours of debugging.

We have already helped dozens of teams streamline their development process. Get a consultation on your environment — it will take no more than an hour.

MLOps: Infrastructure for Training, Deploying, and Monitoring ML Models

The model is trained, metrics — F1 0.94 on validation. Three months later in production, quality drops by 12%. No one knows when — there is no monitoring. It's impossible to retrain quickly — the training script is in a Jupyter notebook of a data scientist who has already left. Data for retraining is collected manually from three disparate systems. About half of the projects come to us with this pain. We build a turnkey MLOps platform: from experiment tracking to automatic deployment and data drift monitoring. We will assess your infrastructure in 1–2 weeks, and in 4–6 weeks you will get a basic MLOps core running in production. Our team has 10+ years of experience in ML infrastructure, over 50 implementations.

How does MLOps infrastructure benefit your ML projects?

Experiment Tracking and Reproducibility

Without tracking, an ML project turns into chaos: it's unclear which checkpoint is better, which hyperparameters were used, which dataset. Reproducing a result a month later is a quest.

Why is experiment tracking the foundation of reproducibility?

MLflow is an open source standard for tracking. It logs parameters, metrics, artifacts (models, graphs), and code. MLflow Model Registry is a centralized model storage with versioning and lifecycle stages (Staging → Production → Archived). Deployment via MLflow Serving or integration with external systems.

Typical initialization in code:

import mlflow

mlflow.set_experiment("fraud-detection-v2")
with mlflow.start_run():
    mlflow.log_params({"learning_rate": 3e-4, "batch_size": 64, "epochs": 10})
    mlflow.log_metric("val_f1", val_f1, step=epoch)
    mlflow.pytorch.log_model(model, "model")

This is the minimum. In production, we add logging of system metrics (GPU utilization, memory), dataset (hash, version), code (git commit hash). Weights & Biases — richer UI, collaboration features, sweep for hyperparameter optimization. MLflow — for on-premise deployment without external dependencies.

DVC (Data Version Control) — versioning of data and models on top of git. Data is stored in S3/GCS/Azure Blob, only metadata (hashes) in git. dvc repro reproduces the entire pipeline from raw data to metrics.

To ensure reproducibility of training, fix random seeds (torch.manual_seed, numpy.random.seed, random.seed) and record them in experiment metadata. Without this, debugging irregular results is painful. Log the dataset version (DVC hash) and git commit — then any experiment can be reproduced down to the byte.

Pipeline Orchestration: Kubeflow, Airflow, Prefect

A pipeline orchestrator becomes necessary when: A 100-line training script in cron is fine for simple tasks. But as soon as you have a multi-step pipeline (data loading → preprocessing → feature engineering → training → validation → deployment if quality above threshold), you need an orchestrator with retry logic, visualization, and alerts.

Kubeflow — Kubernetes-native orchestrator for ML (see Kubeflow). Each step is a Docker container. Supports parallel steps, conditional branches, artifacts between steps. Integrates with Katib (AutoML), KServe (serving), Feast (feature store).

Apache Airflow — more general DAG orchestrator. Wide ecosystem of operators (S3, Spark, DBT, Kubernetes). Easier to deploy if Airflow already exists in the company.

Prefect / Metaflow — less boilerplate. Prefect 2.x with @flow and @task decorators — quick start for small teams.

Typical training pipeline architecture on Kubeflow:

1. Data ingestion component — fetches data from S3/DB, validates schema via Great Expectations
2. Preprocessing component — transformations, normalization, train/val/test split
3. Training component — training on GPU, logging to MLflow
4. Evaluation component — metric calculation, comparison with baseline in Model Registry
5. Conditional deployment — deploy only if new model is better than current by >2% F1

Each component is a separate Docker image. Pipeline is versioned in git. Scheduled run (retraining once a week on new data) or manual.

Model Registry and Lifecycle Management

Model Registry is not just a checkpoint store. It is a centralized system that knows:

• Which model is currently in production (and with what metrics)
• History of all versions with training parameters
• Metadata: dataset, git commit, validation results
• Lifecycle stage: None → Staging → Production → Archived

MLflow Model Registry — standard. For enterprise — Vertex AI Model Registry (GCP), SageMaker Model Registry (AWS), Azure ML Model Registry.

Model promotion through stages: automatically move model to Staging after successful eval, then manual or automatic (during A/B test) promotion to Production. Rollback — switch to previous Production version in seconds.

Serving: From FastAPI to Triton Inference Server

Simple case. FastAPI + PyTorch/ONNX on one server — 80% of production ML deployments are exactly that. Sufficient for most tasks with load up to 100 req/s.

from fastapi import FastAPI
import onnxruntime as ort

app = FastAPI()
session = ort.InferenceSession("model.onnx", providers=["CUDAExecutionProvider"])

@app.post("/predict")
async def predict(request: PredictRequest):
    inputs = preprocess(request.text)
    outputs = session.run(None, {"input_ids": inputs})
    return {"label": postprocess(outputs)}

Triton Inference Server — production standard for high loads (500+ req/s). Dynamic batching, concurrent model execution, model ensemble. Supports TensorRT, ONNX, PyTorch TorchScript, TensorFlow SavedModel.

KServe — Kubernetes-native ML serving with autoscaling, canary deployments, A/B testing out of the box. Scale-to-zero for inactive models — savings on infrastructure up to 40% annually for a project with 10 models.

Monitoring: Data Drift, Model Drift, Infrastructure Metrics

Monitoring — what is usually done last and regretted first. Three levels.

Infrastructure monitoring. Latency (P50/P95/P99), throughput (req/s), error rate (4xx, 5xx), GPU/CPU utilization. Prometheus + Grafana — standard. Alert when P99 latency > threshold or error rate > 1%.

Data drift monitoring. Distribution of input data changes over time. Detect via PSI (Population Stability Index) for numerical features: PSI > 0.2 — strong drift. Chi-squared test for categorical, Kolmogorov-Smirnov test for continuous. Evidently AI — open source library with ready-made drift tests.

Model drift monitoring. If ground truth is delayed (e.g., we know conversion after a week) — monitor real metrics. If not — surrogate metrics: distribution of prediction scores, proportion of confident predictions.

Alerting. Three levels: INFO (minor drift, log it), WARNING (significant, notify team), CRITICAL (quality dropped below threshold — automatic switch to fallback model).

Why is data drift monitoring important?

Without it, you learn about model degradation only from user complaints or ringing SLA. A drift alert allows you to retrain the model in advance, before errors start causing losses. In one of our projects, PSI monitoring detected drift 2 days after a data source change — this saved the campaign.

Feature Store

Feature Store solves the training-serving skew problem. If preprocessing during training and inference is implemented in two different places — divergence is inevitable.

A Feature Store is needed when:

• Several models use the same features
• Features are computed from streaming data (real-time)
• Large team with different people on feature engineering and model training

Feast — open source Feature Store. Offline store (S3 + Parquet) for training, online store (Redis, DynamoDB) for low-latency inference. Feature definitions as code, materialization job syncs offline → online.

Tecton (commercial), Vertex AI Feature Store (GCP), SageMaker Feature Store (AWS) — managed options with less ops overhead.

CI/CD for ML

ML CI/CD is regular CI/CD plus specific ML steps.

ML-specific checks in CI:

• Reproducibility check: run training with a fixed seed, result must match
• Data validation: Great Expectations or Pandera on schema/distribution checks
• Model performance check: automatic eval on holdout, block merge if degradation > threshold
• Latency regression test: inference must meet SLA

GitOps for deployment. Merge to main → CI triggers training → eval → if passes → automatic deployment to Staging → smoke tests → manual promotion to Production or automatic upon successful canary.

Tools: GitHub Actions / GitLab CI for CI, ArgoCD for GitOps deployment on Kubernetes.

What's Included in MLOps Platform Development

We provide a full cycle of work, documentation, and team training.

Total time from audit to full MLOps platform: 3–5 months. Also possible phased launch: basic level (tracking + serving) in 4–6 weeks.

Cost is calculated individually based on data volume, number of models, and infrastructure requirements. Order an MLOps infrastructure audit — get a roadmap in 1–2 weeks. Contact us for a project assessment — we will send a preliminary estimate within 2 business days.

Note: warranty on architectural solutions — 12 months. We provide integration certificates with major cloud providers (AWS, GCP, Azure). During our work, we have not lost a single client after the first implementation — the experience of 50+ successful MLOps projects speaks for itself. Get a consultation on building an MLOps platform today.