Job Url: https://www.indeed.com/viewjob?cmp=Data-Ninjas-Inc.&t=Ai%2Fml+Engineer&jk=907761aea67636d2&xpse=SoA667I3tvuL29S5KZ0LbzkdCdPP&xfps=147ffcb0-5ba9-4f8c-a3ce-816af9fc2d52&xkcb=SoDH67M3tvrtdeyO0B0BbzkdCdPP&vjs=3

Job Description: AI- ML Engineer
Data Ninjas Inc.
Remote
$25 - $30 an hour - Contract

Apply now


 
Profile insights
Here’s how the job qualifications align with your profile.
Skills

Pandas
 (Required)

Natural language processing
 (Required)

Machine learning libraries
 (Required)
+ show more

Do you have experience in Pandas?

Yes

No

Skip
 
Job details
Here’s how the job details align with your profile.
Pay

$25 - $30 an hour
Job type

Contract
 
Full job description
The role

You’ll design, train, and ship compact, reliable models in NLP and/or Computer Vision for real business use cases (e.g., document intelligence for invoices/contracts, OCR, detection/segmentation). You’ll work end-to-end: problem framing → data pipelines → modeling/evaluation → serving and monitoring.

What you’ll do

Translate problems into metrics (F1, ROC-AUC, mAP, IoU, ROUGE) and clear experiment plans.
Build and fine-tune models for text classification, NER, summarization/RAG, OCR, detection/segmentation.
Prototype fast, then productionize: notebooks → services (FastAPI), with tests, Docker, and basic observability.
Own the data loop: labeling strategy, augmentation, drift checks, and error analysis to drive model improvements.
Work across our stack: Hugging Face, PyTorch, OpenCV; IBM for training/inference; watsonx.governance for evals and risk controls; watsonx.data/Databricks for data prep.
Document and demo your work so others can use it (readmes, notebooks, short Looms).
What you’ll bring

1–2 years of hands-on experience (internships and serious portfolio work count) delivering NLP or CV models.
Solid Python; experience with PyTorch (or TensorFlow) and Hugging Face or torchvision.
Practical NLP and/or CV skills (tokenization, LoRA/PEFT, embeddings/RAG, OpenCV, OCR such as Tesseract/TrOCR/Donut).
Comfort serving models (FastAPI/Flask), Docker, Git, and writing clean, testable code.
Data skills: pandas; SQL; plus familiarity with Spark/Databricks or PySpark is a bonus.
Curiosity + product sense: you care about user impact and can explain trade-offs clearly.
Nice to have

Experience with IBM model training/tuning; watsonx.data, watsonx.governance.
Vector databases (pgvector, FAISS, Milvus) and retrieval pipelines.
ML tracking/registry (MLflow or Weights & Biases).
Streamlit/Gradio for rapid prototyping.
Cloud (IBM Cloud Code Engine, AWS, Azure, or GCP); Kubernetes or serverless.
What success looks like (first 90 days)

Ship a small NLP or CV service to staging, with CI and baseline monitoring.
Improve an existing model’s key metric by ~5–10% via targeted data or architecture changes.
Land a clear eval dashboard and an error-analysis workflow the team can repeat.
Our stack

Python, PyTorch, Hugging Face, OpenCV, FastAPI, Docker, Postgres/SQLite, IBM / watsonx.data / watsonx.governance, Spark/Databricks, MLflow/W&B, pgvector/FAISS, GitHub Actions, IBM Cloud Code Engine.

Job Type: Contract

Pay: $25.00 - $30.00 per hour

Expected hours: 40 per week

Work Location: Remote

 
If you require alternative methods of application or screening, you must approach the employer directly to request this as Indeed is not responsible for the employer's application process.

Report job