Podcast
Freelance Data Scientist Playbook: MLOps, Model Monitoring, Upwork & Startup Skills
Open original DataTalks.Club episode
Freelance Data Scientist Playbook: MLOps, Model Monitoring, Upwork & Startup Skills
Original Episode
Use these links for the canonical episode and media sources.
- Open the original DataTalks.Club podcast page
- Watch on YouTube
- Listen on Spotify
- Listen on Apple Podcasts
Episode Overview
How do you transition from startup engineering to a sustainable freelance data science practice while handling MLOps, model monitoring, and client work on Upwork? In this episode, Antonis Stellas — a freelance data scientist at Nanometrisis with a background in applied mathematics, physics and a professional doctorate working on industry consultancy — lays out a practical playbook.
People
Use these links to connect the episode to guest notes.
Chapter Summary
Use these checkpoints to decide whether to open the source transcript.
- 0:00 - Podcast Introduction: guest Antonis and episode themes
- 2:28 - Early Education: applied mathematics, physics and nanotechnology
- 3:50 - Professional Doctorate: industry projects and consultancy in the Netherlands
- 5:35 - Nanometrisis Focus: nanoscale inspection for chips, razors and cosmetics
- 8:19 - Career Choice: choosing a startup over a corporation
- 11:56 - Role Variety in Startups: cross-functional responsibilities
- 14:30 - Small-Team Dynamics: working in a four-person startup
- 15:49 - Skills Acquired: communication, business knowledge and self-organization
- 17:39 - Lean Methodology: build-measure-learn applied to products and ML
- 21:00 - Model Monitoring: data drift, concept drift and Evidently AI
- 21:54 - Community Onboarding: discovering and joining DataTalks.Club
- 25:12 - MLOps Course Project: semiconductor prediction with MLflow, Prefect, Grafana
- 26:43 - Course Recommendations: do exercises, be patient, complete final project
- 28:43 - Open Source Contribution: creating an Evidently how-to during Hacktoberfest
- 30:33 - Starting on Upwork: goals, platform mechanics and client discovery
- 31:57 - Project Types on Upwork: ML, analytics, LLMs and variable durations
- 34:19 - Profile Building: portfolios, attachments and iterative improvements
- 37:09 - Learning from Rejection: refining proposals and specializing skills
- 39:15 - Motivation for Freelancing: learning, extra income and persistence
- 40:39 - Pricing Approach: hourly rates, client type and valuing your time
- 42:33 - Onboarding Workflow: data inspection, milestones and client alignment
- 45:18 - Financial Setup: registering as a freelancer and invoicing considerations
- 47:28 - Balancing Commitments: wearing many hats across startup and freelance work
- 49:40 - Client Acquisition Tips: focus, upskilling and leveraging community resources
- 51:42 - Data Engineering Course: streaming emphasis and Kafka/Confluent usage
- 53:37 - Example Project: streaming YouTube metrics into BigQuery and Looker
- 56:41 - Portfolio Advice: choose projects you enjoy and prioritize exploration
- 58:11 - Recommended Reading: The Lean Startup, Lean Analytics, Designing ML Systems