We use technical cookies and Zorky's de-identified visit log. Yandex Metrica (data is stored on servers in Russia) runs only with your consent. We never sell data. More about cookies and statistics.
Zorky is in test mode — everything is free. Jobs and company pages are open to everyone; links to postings and your personal feed open after you sign up. Sign up →

ML Engineer: market

Data as of November 19, 2026: 2,291 open ML Engineer jobs, 32% remote among jobs that state a work format; median advertised salary $13,666/mo (from 309 postings that state a salary). Source: Zorky, updated daily.

ML Engineer — the largest role inside the AI/ML/DS direction (the biggest pool among ML sub-niches). Production ML — takes a model (own or from a Data Scientist), drives it to a production service with stable SLA. Role family: ML Engineer (general — feature engineering + training + deployment), Senior ML Engineer (end-to-end ML pipelines + scaling), Staff ML Engineer/Principal ML Engineer (ML platform architecture for the whole organisation), ML Tech Lead (team leadership + ML infrastructure decisions), Applied ML Engineer (product ML focused on business metrics). Stack: Python (monopoly), PyTorch/TensorFlow/JAX (deep learning — PyTorch dominates 2026), scikit-learn+XGBoost+LightGBM+CatBoost (classical ML — mandatory), pandas+NumPy+Polars (data manipulation), Hugging Face Transformers+LangChain+LlamaIndex (LLM/NLP), MLflow+Weights & Biases+Neptune (experiment tracking), DVC (data version control), Airflow/Prefect/Dagster (orchestration), Docker+Kubernetes (deployment), FastAPI+Triton Inference Server+BentoML+TorchServe (model serving), ONNX (cross-framework inference), Feast/Tecton (feature stores), Spark+Ray+Dask (distributed compute), AWS SageMaker/GCP Vertex AI/Azure ML (cloud-managed). Top stack: python, cloud, pytorch, llm, aws. Among jobs with a stated work format, 32% are remote.

Data as of:

Key numbers for the ML Engineer market: open jobs, median salary, observed talent pool and remote share.

2,291
open jobs — by job title, reposts merged
$13,666
median $/mo
32%
remote

The ML Engineer market currently has 2,291 open roles, 2,024 of them freshly observed. Median salary $13,666/mo.

Among jobs with a stated work format, 32% are remote. ML work is cloud-based (training on cloud GPUs, serving in K8s). Outsourcers (EPAM AI / Luxoft AI) — almost always remote. Russian banks (Sber AI banking ML) — hybrid/office compliance. International AI companies — full-remote standard.

⚠ salary known for 309 of 2291 jobs; remote share from 1926 with a stated format; 2024 counted as fresh observations; trend and hiring difficulty are not shown

Timeline: postings and sources

How many ML Engineer postings appeared each week since Jun 1, and from how many sources. The lower chart separates market growth from coverage growth: a jump in the week new sources were connected means we started seeing more, not that the market grew.

New postings per week

Chart data
New postings per week. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodNew postings per week
2026-06-01104
2026-06-08142
2026-06-1577
2026-06-22175
2026-06-29332
2026-07-06286
2026-07-13353
2026-07-20283
2026-07-27170
2026-08-03215
2026-08-10128
2026-08-17115
2026-08-24132
2026-08-31115
2026-09-07127
2026-09-14186
2026-09-21612
2026-09-28370
2026-10-05373
2026-10-12794
2026-10-19492
2026-10-26519
2026-11-02406
2026-11-09463

Active sources per week

Sources over the period: 217. ▲ — a week with a large batch of new sources.

Chart data
Active sources per week. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodActive sources
2026-06-0117
2026-06-0823
2026-06-1515
2026-06-2232
2026-06-2953
2026-07-0644
2026-07-1354
2026-07-2057
2026-07-2746
2026-08-0346
2026-08-1031
2026-08-1729
2026-08-2435
2026-08-3140
2026-09-0741
2026-09-1435
2026-09-2156
2026-09-2852
2026-10-0561
2026-10-1263
2026-10-1966
2026-10-2666
2026-11-0263
2026-11-0976
New sources connected: Jun 29 +30 · Jul 6 +18 · Jul 13 +18 · all weeks (23) in the table
WeekNew sourcesNew postings
Jun 8+11142
Jun 15+277
Jun 22+15175
Jun 29+30332
Jul 6+18286
Jul 13+18353
Jul 20+11283
Jul 27+8170
Aug 3+2215
Aug 10+7128
Aug 17+2115
Aug 24+5132
Aug 31+9115
Sep 7+4127
Sep 14+4186
Sep 21+10612
Sep 28+3370
Oct 5+10373
Oct 12+3794
Oct 19+9492
Oct 26+4519
Nov 2+5406
Nov 9+10463

The current week is not over yet and is not shown: an incomplete week would look like a drop.

Week = posting first-seen date, all statuses; postings are selected by the same rule as this page's counter. Sources are distinct collection channels; their names are not published.

Chart data
Timeline: postings and sources — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
WeekNew postings per weekActive sources
2026-06-0110417
2026-06-0814223
2026-06-157715
2026-06-2217532
2026-06-2933253
2026-07-0628644
2026-07-1335354
2026-07-2028357
2026-07-2717046
2026-08-0321546
2026-08-1012831
2026-08-1711529
2026-08-2413235
2026-08-3111540
2026-09-0712741
2026-09-1418635
2026-09-2161256
2026-09-2837052
2026-10-0537361
2026-10-1279463
2026-10-1949266
2026-10-2651966
2026-11-0240663
2026-11-0946376

Salaries over time

Median ML Engineer salary per week, USD per month.

Median salary, USD per month

Chart data
Median salary, USD per month. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodMedianPostings with a salary per week (n)
2026-09-2112,54516
2026-09-2810,23828
2026-10-059,45739
2026-10-129,80556
2026-10-1913,91992
2026-10-2613,13493
2026-11-0213,75085
2026-11-0912,929110

The series starts on Sep 21: there is no data before that.

Postings with a salary per week (n)

Only postings with a USD salary and a known period (month or year), $500–$50,000 per month. Weeks with fewer than 10 such postings are left blank — that means no data, not zero.

Chart data
Salaries over time — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
WeekMedian
2026-06-01—
2026-06-08—
2026-06-15—
2026-06-22—
2026-06-29—
2026-07-06—
2026-07-13—
2026-07-20—
2026-07-27—
2026-08-03—
2026-08-10—
2026-08-17—
2026-08-24—
2026-08-31—
2026-09-07—
2026-09-14—
2026-09-2112,545
2026-09-2810,238
2026-10-059,457
2026-10-129,805
2026-10-1913,919
2026-10-2613,134
2026-11-0213,750
2026-11-0912,929

Remote work over time

Share of ML Engineer postings offering fully remote work, per week.

Remote postings, %

Chart data
Remote postings, %. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodRemote
2026-06-0144.4
2026-06-0819.1
2026-06-1511.5
2026-06-2222
2026-06-2925.5
2026-07-0634.2
2026-07-1328.4
2026-07-2047.3
2026-07-2736.5
2026-08-0331.2
2026-08-1034.3
2026-08-1741.8
2026-08-2441.7
2026-08-3134.6
2026-09-0728.4
2026-09-1425
2026-09-2128
2026-09-2835.7
2026-10-0537.6
2026-10-1238.8
2026-10-1937.8
2026-10-2646.9
2026-11-0235.1
2026-11-0935.3

The share is among postings that state a format (remote, hybrid, office). Weeks with fewer than 10 such postings are left blank.

Chart data
Remote work over time — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
WeekRemote
2026-06-0144.4
2026-06-0819.1
2026-06-1511.5
2026-06-2222
2026-06-2925.5
2026-07-0634.2
2026-07-1328.4
2026-07-2047.3
2026-07-2736.5
2026-08-0331.2
2026-08-1034.3
2026-08-1741.8
2026-08-2441.7
2026-08-3134.6
2026-09-0728.4
2026-09-1425
2026-09-2128
2026-09-2835.7
2026-10-0537.6
2026-10-1238.8
2026-10-1937.8
2026-10-2646.9
2026-11-0235.1
2026-11-0935.3

Technologies over time

How the share of ML Engineer postings mentioning each of the top technologies changed.

Postings mentioning the technology, %

Chart data
Postings mentioning the technology, %. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodPythoncloudllm
2026-08-10402017.8
2026-08-1737.726.430.2
2026-08-24322025.3
2026-08-3148.123.126.9
2026-09-0734.513.829.3
2026-09-1437.226.921.8
2026-09-2155.536.132.3
2026-09-2859.933.728.2
2026-10-0563.139.735.9
2026-10-1255.635.428.6
2026-10-196341.244.3
2026-10-2666.637.935.1
2026-11-026529.135.6
2026-11-0968.937.736.6
Up to 5 on the chart:

Share among postings with a tagged stack; the series starts on Aug 10: earlier stack tags were assigned differently. Weeks with fewer than 20 tagged postings are left blank.

Chart data
Technologies over time — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
WeekPythoncloudllmpytorchAWS
2026-08-10402017.88.98.9
2026-08-1737.726.430.2179.4
2026-08-24322025.318.76.7
2026-08-3148.123.126.917.311.5
2026-09-0734.513.829.320.713.8
2026-09-1437.226.921.819.26.4
2026-09-2155.536.132.332.916
2026-09-2859.933.728.236.922.6
2026-10-0563.139.735.933.827.5
2026-10-1255.635.428.632.220.2
2026-10-196341.244.334.326.7
2026-10-2666.637.935.132.927.2
2026-11-026529.135.637.416.3
2026-11-0968.937.736.636.329.5

Countries over time

How the share of ML Engineer postings in each of the top countries changed, per week.

Postings in the country, %

Chart data
Postings in the country, %. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodUSAUKSingapore
2026-06-0164.620
2026-06-0868.90.83.8
2026-06-1565.85.51.4
2026-06-2233.11.218.4
2026-06-2918.20.747.9
2026-07-0652.26.56.1
2026-07-1345.63.311.8
2026-07-2038.88.615.5
2026-07-2750.7311.9
2026-08-0340.215.66.1
2026-08-1038.214.72.9
2026-08-173814.10
2026-08-2433.624.30
2026-08-3121.1301.1
2026-09-0730.928.72.1
2026-09-143827.22.5
2026-09-2143.6147.5
2026-09-2830.219.96.9
2026-10-0531.818.24.5
2026-10-1232.111.22
2026-10-1929.114.34.4
2026-10-2629.2123.9
2026-11-0235.9148.6
2026-11-0944.9112.2
Up to 5 on the chart:

Share among postings with a known country; the chart shows the five countries with the most postings over six months. Weeks with fewer than 20 postings with a known country are left blank. A rising share can mean a newly connected source in that country rather than market growth.

Chart data
Countries over time — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
WeekUSAUKSingaporePolandIndia
2026-06-0164.62016.26.1
2026-06-0868.90.83.86.86.8
2026-06-1565.85.51.46.82.7
2026-06-2233.11.218.421.53.7
2026-06-2918.20.747.94.35.9
2026-07-0652.26.56.14.94.9
2026-07-1345.63.311.83.69.2
2026-07-2038.88.615.513.55.7
2026-07-2750.7311.910.44.5
2026-08-0340.215.66.14.55.6
2026-08-1038.214.72.929.8
2026-08-173814.1004.3
2026-08-2433.624.303.78.4
2026-08-3121.1301.11.18.9
2026-09-0730.928.72.12.19.6
2026-09-143827.22.53.85.7
2026-09-2143.6147.52.24.9
2026-09-2830.219.96.98.56.9
2026-10-0531.818.24.59.96.4
2026-10-1232.111.228.28
2026-10-1929.114.34.49.96.2
2026-10-2629.2123.95.67.2
2026-11-0235.9148.646
2026-11-0944.9112.26.84.9

Clean market: hiring index

The same 7,505 employer job boards; the posting count comes from the board itself. Connecting new sources does not move this index — it shows the market, not our coverage. This is the whole panel market, not just ML Engineer.

Hiring index, Oct 16 = 100

Chart data
Hiring index, Oct 16 = 100. Block data table. A missing value is not zero; units and sample are stated in the metric description.
PeriodHiring index
2026-10-16100
2026-10-17100.03
2026-10-1899.76
2026-10-1999.68
2026-10-2099.27
2026-10-2198.6
2026-10-2299.07
2026-10-2397.4
2026-10-2497.41
2026-10-2697.33
2026-10-2796.76
2026-10-2896.21
2026-10-2995.17
2026-10-3095.13
2026-10-3193.65
2026-11-0193.33
2026-11-0292.67
2026-11-0392.65
2026-11-0492.41
2026-11-0592.13
2026-11-0690.44
2026-11-0790.48
2026-11-0890.45
2026-11-1090.29
2026-11-1190.22
2026-11-1390.12
2026-11-1489.7
2026-11-1589.68
2026-11-1689.62
2026-11-1888.91
2026-11-1988.02

Chain index: each day compares only boards polled on both that day and the previous one; outliers and platform-wide polling outages are excluded. Days with no comparable pairs are left blank.

Chart data
Clean market: hiring index — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
DayHiring index
2026-10-16100
2026-10-17100.03
2026-10-1899.76
2026-10-1999.68
2026-10-2099.27
2026-10-2198.6
2026-10-2299.07
2026-10-2397.4
2026-10-2497.41
2026-10-2697.33
2026-10-2796.76
2026-10-2896.21
2026-10-2995.17
2026-10-3095.13
2026-10-3193.65
2026-11-0193.33
2026-11-0292.67
2026-11-0392.65
2026-11-0492.41
2026-11-0592.13
2026-11-0690.44
2026-11-0790.48
2026-11-0890.45
2026-11-1090.29
2026-11-1190.22
2026-11-1390.12
2026-11-1489.7
2026-11-1589.68
2026-11-1689.62
2026-11-1888.91
2026-11-1988.02

How many open ML Engineer jobs there are and how many matching specialists Zorky observes.

Demand and observed supply

Chart data
Block data table. A missing value is not zero; units and sample are stated in the metric description.
MetricValue
Open demand2,291
Open demand2,291
Observed supply— not published

ML Engineer — the largest AI/ML role, demand growth 2026 driven by the LLM/Generative AI boom: Foundation Model deployment, RAG applications, AI agents, voice/multimodal products. Russian product companies — a steady flow. EPAM/Luxoft AI Practice — the largest outsource channel for US AI projects.

⚠ candidates matching a vacancy are not counted yet: demand and the observed pool are shown

How salaries in ML Engineer jobs are distributed, in US dollars per month.

Salary distribution

P25 $9,600 · P50 $13,666 · P75 $17,416
Chart data
Block data table. A missing value is not zero; units and sample are stated in the metric description.
MetricValue
<$1k4
$1–2k2
$2–3k7
$3–4k7
$4–5k9
$5–6k14
$6–7k11
$7–8k4
$8–9k9
$9–10k16
$10–12k31
$12–15k68
$15–20k83
$20k+44
Salary in postings: base 2,291
  • usable for the calculation: 309 (13.5%)
  • stated but not converted to USD/month: 372 (16.2%)
  • not stated: 1,610 (70.3%)

One in ten postings offers under $5,027, one in ten over $20,916. Half of the postings offer between $9,600 and $17,416. Sample: 309.

Chart data
Salary distribution — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
MetricLower boundUpper boundCount
<$1k01,0004
$1–2k1,0002,0002
$2–3k2,0003,0007
$3–4k3,0004,0007
$4–5k4,0005,0009
$5–6k5,0006,00014
$6–7k6,0007,00011
$7–8k7,0008,0004
$8–9k8,0009,0009
$9–10k9,00010,00016
$10–12k10,00012,00031
$12–15k12,00015,00068
$15–20k15,00020,00083
$20k+20,000—44

Which countries have the most open ML Engineer jobs.

Demand geography

countryjobsshare
United States88245.9%
United Kingdom1608.3%
India1206.2%
Poland1176.1%
Singapore864.5%
Canada633.3%
Germany583%
France382%
Russia341.8%
Netherlands261.4%
Other countries33817.6%

The leader by ML Engineer job count is the US (882 positions). Poland — AI-friendly EU hub (Warsaw/Krakow). Germany — Berlin AI-startup cluster.

⚠ job counts only: salary by country is not published

Chart data
Demand geography — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
CountryJobs
US882
GB160
IN120
PL117
SG86
CA63
DE58
FR38
RU34
NL26

Which technologies are most often required in the same job as ML Engineer.

Used together with

python
93540.8%
cloud
57225%
llm
52122.7%
pytorch
51422.4%
aws
38416.8%
mlops
31313.7%
tensorflow
25911.3%
docker
25811.3%
azure
25010.9%
fine-tuning
2259.8%

Share — how many of the 2,291 jobs in the ML Engineer market mention this technology. Ordered by job count, not by “strength of association”. Umbrella terms like cloud are not a specific cloud (AWS, Azure).

Top ML Engineer stack 2026: Python (monopoly), PyTorch / TensorFlow / JAX (PyTorch dominates), scikit-learn + XGBoost + LightGBM + CatBoost (classical), pandas + NumPy + Polars (data), Hugging Face + LangChain + LlamaIndex (LLM/NLP), sentence-transformers + Pinecone / Weaviate / Qdrant / Milvus (vector DBs / RAG), MLflow + W&B + Neptune (experiment tracking), DVC, Airflow / Prefect / Dagster, Docker + Kubernetes, FastAPI + Triton Inference Server + BentoML + TorchServe + vLLM / TGI / SGLang (LLM serving), ONNX, Feast / Tecton (feature stores), Ray + Spark + Dask, AWS SageMaker / GCP Vertex AI / Azure ML, Optuna, Evidently AI (drift). Senior add-ons: LoRA / QLoRA fine-tuning, quantization, distillation.

Common pairs: Python + PyTorch + Hugging Face, Python + scikit-learn + XGBoost + LightGBM, MLflow + DVC + Airflow, Docker + Kubernetes + Triton Inference Server, FastAPI + Triton + ONNX, LangChain + Pinecone + OpenAI API, Ray + PyTorch DDP (distributed training), Feast + Spark + Snowflake. Learning roadmap: Python + math → classical ML (sklearn / XGBoost) → Deep Learning (PyTorch) → MLOps (MLflow + Docker + K8s) → LLM specialisation (Hugging Face + LangChain + RAG) → distributed training (Ray) → cloud-managed ML (SageMaker / Vertex AI).

Chart data
Used together with — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
NameCount
python935
cloud572
llm521
pytorch514
aws384
mlops313
tensorflow259
docker258
azure250
fine-tuning225
kubernetes225
gcp181

How many ML Engineer jobs there are at each level, from junior to principal.

Demand by grade

gradejobs
senior794
principal341
junior102
lead90
middle87

Junior — rare, typical entry via Data Analyst → ML Junior or Backend Middle → ML Junior. Career flow: Junior (1-2 years) → Middle (2-3 years) → Senior → either Staff / Principal ML Engineer (deep technical), ML Tech Lead / Engineering Manager, a move into Research (with PhD), or ML CTO/Founder at an AI startup.

⚠ demand side only: the grade of the observed pool is unknown for most of it

Chart data
Demand by grade — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
LevelCount
senior794
principal341
junior102
lead90
middle87

Who hires the most for ML Engineer and how concentrated demand is among the largest companies.

Employers

Demand is spread across 1,312 employers. The largest accounts for 1.7%, the top ten for 8.6%; the remaining 91.4% is long tail.

largest 1.7%2nd–3rd 2.2%4th–10th 4.7%the rest 91.4%
sharevalue
top-11.7%
top-33.9%
top-108.6%
long tail91.4%

⚠ names are not shown: staffing agencies and end employers are not yet told apart by the classifier

Which company sites and job boards Zorky collects ML Engineer jobs from.

Where Zorky sees this market

Observed across 109 sources; the largest accounts for 20.7% — this market does not rest on a single channel.

ML Engineer jobs: hh.ru, Habr Career, getmatch, Djinni, LinkedIn (huge international AI segment), NoFluffJobs / JustJoin.it (Poland AI-friendly), Telegram, career pages of EPAM AI / Luxoft AI / Andersen AI, specialised boards aijobs.net + ai-jobs.net + builtin.com/jobs/ai, Y Combinator Work at a Startup, AI-startup careers pages.

Chart data
Where Zorky sees this market — ML Engineer. Block data table. A missing value is not zero; units and sample are stated in the metric description.
ClassNameEnglish nameCountShare
atsATS и карьерные сайтыATS & career sites1,5460.6748
boardджоб-борды и агрегаторыjob boards & aggregators6160.2689
telegramTelegramTelegram1150.0502
communityсообщества и публичные профилиcommunity & public profiles140.0061

The newest ML Engineer jobs from company career sites.

Recent openings

All jobs →

Latest open ML Engineer jobs — the most recent 10 positions with adequate description quality. The full list is in our CRM or via the "see all" link below.

How this is measured
Vacancy
an open job that cleared the quality gate and lists at least two technologies
Observed candidate
a candidate whose stack contains this technology; an aggregate — not a single record leaves the perimeter
Matchable candidate
not counted yet
Window
jobs open at the moment the snapshot was built

About the data

  • Some breakdowns are hidden: their data coverage is not yet sufficient.
  • Statistics are shown only where the sample clears a quality gate.
  • A missing block does not mean a value of zero.

Breakdowns currently hidden: 8.

Direction: AI / ML / Data Science

Related specializations

Inside ML Engineer: general ML Engineer (universal, recsys/NLP/CV/fraud), Applied ML Engineer (product focus + business metrics), Senior ML Engineer (end-to-end pipelines + scaling), Staff / Principal ML Engineer (ML platform architecture for the organisation), ML Tech Lead (team leadership), ML Platform Engineer (infrastructure for other ML teams). Comparison with data-scientist/research/mlops — in the bar below.

AI / LLM EngineerComputer VisionData ScientistMLOpsNLP EngineerResearch Engineer

Frequently asked questions

The most common questions about ML Engineer: pay, ML Engineer vs DS vs Research vs MLOps, production ML pipeline stack (10 components), how it differs from Applied ML, remote, how to start (8-14 months from Data Analyst / Backend Middle), Senior skills (LLM mastery + MLOps + distributed). Answers recompute automatically.

What does an ML Engineer Junior, Middle, Senior, or Lead earn?

Junior ML Engineer — rare (the market starts with Data Analyst or Backend → ML). The Junior → Middle jump — after mastering the first production ML pipeline end-to-end (feature engineering → training → deployment → monitoring). Middle → Senior — multi-pipeline ownership + LLM/RAG experience + scaling distributed training + cost optimisation. Senior → Staff/Principal — ML platform architecture, multi-team ML strategy. Career flow: Data Analyst (1-2 years) → Junior ML Engineer (1-2 years) → Middle (2-3 years) → Senior → either Staff ML Engineer (deep technical), ML Tech Lead (management), a move into Research (PhD track), or ML Founder/CTO at a startup.

What stack does an ML Engineer most often need?

Top 5: python, cloud, pytorch, llm, aws. Python is the monopoly (R is barely seen in production ML 2026). Classical ML: scikit-learn + XGBoost + LightGBM + CatBoost — base, must for tabular data. Data manipulation: pandas + NumPy + Polars (modern, rising 2026 — 10-100× faster), PyArrow for cross-framework data. Computer Vision: OpenCV, torchvision, Detectron2, MMDetection, Ultralytics YOLO. Experiment tracking: MLflow (open-source standard) or Weights & Biases (premium UI) or Neptune. Data version control: DVC. Orchestration: Airflow + Prefect + Dagster (one deeply). Deployment: Docker + Kubernetes mastery, FastAPI + Triton Inference Server + BentoML + TorchServe (one deeply). Cross-framework inference: ONNX. Feature stores: Feast (open-source) or Tecton (managed). Distributed compute: Spark + Ray + Dask. Cloud-managed ML: AWS SageMaker / GCP Vertex AI / Azure ML (one deeply). Senior extras: quantization (PyTorch quantization, bitsandbytes), distillation, LoRA/QLoRA fine-tuning, vLLM/TGI/SGLang for LLM serving, Optuna for hyperparameter tuning.

ML Engineer vs Data Scientist vs Research vs MLOps — what's the difference?

Data Scientist — focus on business problem framing + statistical modelling + A/B testing + insights communication. Stack: pandas + sklearn + Jupyter + SQL + Tableau/Looker. Less infrastructure-heavy. See Data Scientist. ML Engineer (this page) — focus on production ML pipelines + deployment + scaling. Stack like DS + Docker/K8s + MLflow + Airflow + cloud-managed ML. Research Engineer / Scientist — focus on new models, papers, novel architectures. Often PhD track. Stack: PyTorch deep + JAX + experiment-heavy infrastructure. See Research Engineer / Scientist. MLOps Engineer — focus on ML platform infrastructure (feature stores, model registry, monitoring, observability, cost optimisation). Stack: K8s + Terraform + MLflow/Kubeflow + Prometheus/Grafana + cloud. See MLOps Engineer. Career switch ML Engineer ↔ DS — 1-3 months. ML Engineer ↔ MLOps — 2-4 months (if K8s base exists). Research path requires PhD or 3+ years of deep specialisation.

What does the production ML pipeline stack look like in 2026?

Reference stack for production ML 2026: 1) Data ingestion — Airflow/Prefect/Dagster pulls raw data from source systems (DB/Kafka/S3). 2) Feature engineering — Spark/Pandas/Polars transforms → Feature Store (Feast or Tecton — centralised storage, point-in-time correctness, online+offline serving). 3) Experimentation — Jupyter + W&B/MLflow tracking + DVC for data version control + Optuna for hyperparameter search. 4) Training — Ray Train or PyTorch Lightning + distributed (multi-GPU/multi-node), checkpointing in S3/GCS, quantization (INT8/FP16) for inference cost. 5) Model registry — MLflow Model Registry or SageMaker Model Registry (versioning + stages: staging/production/archived). 6) Validation — automated tests (data drift, prediction quality), shadow deployment, A/B test setup. 7) Serving — Triton Inference Server (NVIDIA, multi-framework + dynamic batching) or BentoML or TorchServe or vLLM (for LLM). FastAPI wrapper for API contract. 8) Deployment — Kubernetes (production standard) + Helm + ArgoCD for GitOps. 9) Monitoring — Prometheus + Grafana (latency/throughput/error rate) + custom drift detection (Evidently AI/WhyLabs) + cost dashboards. 10) Feedback loop — log predictions + ground truth → retraining trigger. ML platform for the whole organisation — Kubeflow or Metaflow or SageMaker Pipelines. LLM-specific stack: vLLM/TGI/SGLang for serving + prompt engineering (LangChain/LlamaIndex) + RAG (vector DBs Pinecone/Weaviate/Qdrant) + fine-tuning via LoRA/QLoRA + evaluation (RAGAS, DeepEval, custom benchmarks).

What share of jobs are remote?

Among jobs with a stated work format, 32% are remote. Jobs without a stated format are excluded from the denominator; hybrid work is not counted as remote.

How is ML Engineer different from Applied ML Engineer?

ML Engineer (general) — can work in any ML domain (recsys, NLP, CV, fraud detection, ranking). Stack is universal. Applied ML Engineer — focus on a specific product ML task + business-metrics ownership (CTR, conversion, revenue lift). Stack is the same, but adds product thinking + A/B testing mastery + stakeholder communication. Pay comparable (median), but at product companies Applied ML is usually somewhat higher at Senior thanks to business-impact ownership. ML Platform Engineer (a separate sub-niche — closer to MLOps) — builds infrastructure for other ML Engineers (Feature Store, MLflow infrastructure, training platform). Career choice: ML Engineer if universality and portability matter most, Applied ML Engineer if product impact is interesting, ML Platform Engineer if infrastructure and tooling. All three are legitimate Senior+ paths, not better/worse.

Where to start in ML Engineering in 2026?

Roadmap: 1) Python deep to Backend Middle level (clean code, design patterns, async). Books: "Fluent Python" Ramalho, "Effective Python" Slatkin. 2) Math — linear algebra + calculus + probability/statistics at the level of "understanding what's under the hood of backprop". MIT 18.06 Linear Algebra (Strang), 3Blue1Brown "Essence of Linear Algebra". 3) Classical ML — scikit-learn + XGBoost + LightGBM mastery. Andrew Ng "Machine Learning Specialization" (Coursera). Book "Hands-On Machine Learning" Géron (must-read 2026). 4) Deep Learning — PyTorch deep (forward/backward/optimisers/training loops from scratch). "Deep Learning Specialization" Andrew Ng. "Deep Learning with PyTorch" Stevens. 5) NLP/LLM track: Hugging Face course (free) → fine-tuning small LLMs (Mistral 7B/Llama 3 8B) with LoRA/QLoRA → build a RAG application with LangChain + Qdrant. 6) Computer Vision track (alternative): torchvision + Detectron2 + Ultralytics YOLO + custom dataset training. 7) MLOps basics: MLflow + DVC + Airflow + Docker + Kubernetes for ML serving. "Designing Machine Learning Systems" Chip Huyen — must-read for production ML. 8) Cloud ML: one of AWS SageMaker / GCP Vertex AI / Azure ML — go through one e2e tutorial. 9) Distributed training: Ray Train + PyTorch DDP. 10) Pet project: full e2e ML system (data pipeline + feature store + training + MLflow tracking + Kubernetes deployment + monitoring) on an open dataset (Kaggle). International (EN): fast.ai "Practical Deep Learning", Hugging Face NLP Course, DeepLearning.AI (Andrew Ng). Classic books: "Pattern Recognition and Machine Learning" Bishop, "The Hundred-Page Machine Learning Book" Burkov. Conferences: NeurIPS, ICML, ICLR (papers — must reading for Senior). Data Analyst or Backend Middle → ML Engineer Junior — 8-14 months.

How many open jobs are in this sample?

The published sample contains 2291 open jobs after merging reposts. This is the same count shown in the main indicator. The geography section describes the sample coverage; this is not an estimate of the entire market.

What skills does a Senior ML Engineer need?

A Senior ML Engineer owns the full production ML cycle + technical leadership. Programming mastery: Python Backend Senior level (clean code, async, profiling, memory optimisation). Math foundations: linear algebra + calculus + probability at the level of "reading papers without blocks". Classical ML mastery: deep feature engineering, model selection rationale, calibration, evaluation metrics trade-offs, business-aligned model design. Deep Learning mastery: PyTorch deep (custom Layers/Modules/Optimizers), distributed training (DDP, FSDP), mixed-precision, quantization (INT8/FP16), distillation, ONNX export. LLM mastery (2026 must for Senior): fine-tuning via LoRA/QLoRA, prompt engineering, RAG architecture (vector DBs Pinecone/Weaviate/Qdrant), evaluation (RAGAS, DeepEval), vLLM/TGI/SGLang for serving. MLOps: MLflow / W&B mastery, DVC, Airflow / Prefect / Dagster, Feature Stores (Feast / Tecton), Docker + Kubernetes mastery for ML serving, Helm + ArgoCD. Cloud ML: one of AWS SageMaker / GCP Vertex AI / Azure ML deeply. Distributed compute: Ray + Spark mastery. Monitoring: Prometheus + Grafana + Evidently AI (drift detection) integration. Cost optimisation: GPU utilisation, batch sizing, model quantization, caching strategies. System design for ML: design ML systems on whiteboard at 100M+ user scale (recsys / fraud / ranking / chatbot). Soft: papers reading (NeurIPS / ICML / ICLR), technical writing (ADRs, design docs), code review for multiple teams, mentoring Middle ML Engineers, communication with product/data/backend teams, A/B test design + statistical analysis. English for Senior+ MUST — the ML/AI community and papers are in English. Bonus: open-source contributions (PyTorch / Hugging Face / LangChain) — sharply increase market value.

Leave a request

Describe the task and leave a contact — the request goes to our CRM and we reply at the contact you provide.

By submitting, you agree to the privacy policy.