Roles · AI platform engineering
LLM and MLOps engineers for AI products and model platforms
LLM engineers build products on large language models, such as AI agents, retrieval-augmented generation and structured extraction. MLOps engineers build the platform that trains, serves, monitors and retrains models in production. AI labs, fintechs and software companies hire both. Kernavique fills these roles two ways: Capacity places engineers on contract through their own companies, and Search makes permanent hires.
Last updated
LLM and MLOps engineer
What the role does
- Builds products on large language models end to end: AI agents, retrieval-augmented generation, structured extraction and fine-tuning, with the infrastructure for production LLM inference.
- Designs shared AI platforms that other teams plug into, such as a reusable agent platform or a company-wide LLM gateway with unified APIs, rate and token limits and failover.
- Builds the infrastructure behind feature engineering, model training and model serving, including inference for predictive models and LLMs, and streaming feature pipelines.
- Automates the pipelines that build, deploy, run, monitor, debug and retrain models, with CI/CD and automated testing for model deployment.
- Sets up evaluation, tracing and observability so that non-deterministic systems can be measured and governed in production.
- Runs distributed training and batch inference on Ray or Spark, and shares on-call for the platform.
Skills and credentials
Agents and retrieval
Ramp’s Applied AI Engineer ad covers AI agents, retrieval-augmented generation, structured extraction and fine-tuning, and asks for full-stack engineers comfortable with web frameworks, backend and infrastructure.
Evaluation and tracing
Brex prefers experience with human-in-the-loop workflows, tracing, evaluation and feedback mechanisms. Chime’s ML platform ad asks for evaluation frameworks built for non-deterministic systems.
LLM gateways and model access
Reddit’s GenAI platform ad centers on an LLM gateway: unified APIs for internally and externally hosted LLMs, rate and token-limit management, and failover. Chime lists model access, prompt lifecycle, and retrieval and tool integration.
Feature stores, training and serving
Coinbase asks for 5+ years and familiarity with feature stores, serving frameworks and training orchestration. Gusto asks for 5+ years of software engineering and platform services for the ML lifecycle, such as feature stores, deployment and observability.
CI/CD and infrastructure as code
Gusto wants CI/CD and automated testing for model deployment, and prefers AWS. Chime adds CI/CD for models, Terraform and on-call for an ML and AI platform that runs on AWS.
Distributed compute, streaming and GPUs
Chime’s ad lists distributed training and batch inference on Ray or Spark, a feature store, Kafka, Kinesis or Flink streaming, and GPU workloads as a nice-to-have, with CUDA a plus.
What the market pays
- OpenAI
$251K – $278K • Offers Equity
Applied AI Engineer, Enterprise
San Francisco. Not stated.
- Ramp
$204.4K – $352K • Offers Equity
Applied AI Engineer
New York, NY; San Francisco, CA. Not stated.
- Brex
$216,000 - $240,000 USD
AI Engineer, Workbench
San Francisco, California, United States. Not stated.
- Writer
SF & NYC: $146K – $240K. Other US locations: $132K – $216K
Software engineer, generative AI
San Francisco, CA; Seattle, WA; New York City, NY. Base salary, plus equity.
- Reddit
$292,500 – $409,500 USD
Senior Staff Machine Learning Engineer, GenAI Platform
Remote - United States. Base salary.
- Airbnb
$292,000 — $365,000 USD
Principal Machine Learning Engineer- LLM Fine-tuning and Optimization
United States (US - Remote Eligible). Base salary.
- Coinbase
$191,250 — $225,000 USD
Senior Machine Learning Platform Engineer
Remote - USA. Base salary.
- Chime
$187,000 – $259,000 USD
Senior Software Engineer, Machine Learning Platform
San Francisco, CA, USA. Base salary.
- Gusto
Denver: $160,000–$200,000/year. San Francisco and New York: $190,000–$240,000/year
Software Engineer, ML Platform
Denver, CO; New York, NY; San Francisco, CA. Not stated.
- Samsara
$200,200 — $357,500 USD
Lead Machine Learning Engineer - ML Infrastructure
Remote - US. Base salary.
| Employer | Role, as titled in the ad | Location | Published range | Basis |
|---|---|---|---|---|
| OpenAI | Applied AI Engineer, Enterprise | San Francisco | $251K – $278K • Offers Equity | Not stated |
| Ramp | Applied AI Engineer | New York, NY; San Francisco, CA | $204.4K – $352K • Offers Equity | Not stated |
| Brex | AI Engineer, Workbench | San Francisco, California, United States | $216,000 - $240,000 USD | Not stated |
| Writer | Software engineer, generative AI | San Francisco, CA; Seattle, WA; New York City, NY | SF & NYC: $146K – $240K. Other US locations: $132K – $216K | Base salary, plus equity |
| Senior Staff Machine Learning Engineer, GenAI Platform | Remote - United States | $292,500 – $409,500 USD | Base salary | |
| Airbnb | Principal Machine Learning Engineer- LLM Fine-tuning and Optimization | United States (US - Remote Eligible) | $292,000 — $365,000 USD | Base salary |
| Coinbase | Senior Machine Learning Platform Engineer | Remote - USA | $191,250 — $225,000 USD | Base salary |
| Chime | Senior Software Engineer, Machine Learning Platform | San Francisco, CA, USA | $187,000 – $259,000 USD | Base salary |
| Gusto | Software Engineer, ML Platform | Denver, CO; New York, NY; San Francisco, CA | Denver: $160,000–$200,000/year. San Francisco and New York: $190,000–$240,000/year | Not stated |
| Samsara | Lead Machine Learning Engineer - ML Infrastructure | Remote - US | $200,200 — $357,500 USD | Base salary |
These are ranges for permanent roles, copied exactly from employers’ own ads, and “Not stated” means the ad does not say whether its range is base. Arrangements for engineers on contract through their own companies vary by engagement. Each employer links to its ad. All read October 11, 2026.
How we fill it
- Platform engineers
Capacity
LLM, GenAI and MLOps engineers on contract through their own companies, each matched to a written scope and term before they start.
- Engineering search
Search
Permanent LLM, applied AI and ML platform engineers, hired against a specification we write back to you after the brief call.
- Contract to permanent
Capacity
Start an engineer on contract on a defined piece of your platform, and convert them to a permanent employee at the end of the term if both sides want it.
- Capability Blueprint
Advisory
For a first AI platform team: which LLM and MLOps roles to hire, in what order, and which to hold on contract or permanently.
Questions buyers ask
- What is the difference between an LLM engineer and an MLOps engineer?
- An LLM engineer builds the product: Ramp’s Applied AI Engineer ships AI agents, retrieval-augmented generation and structured extraction end to end. An MLOps engineer builds the platform underneath: Gusto’s ML Platform engineer automates the pipelines that build, deploy, monitor and retrain models. The two meet in platform teams, and Chime’s Machine Learning Platform ad covers model access, prompt lifecycle and evaluation for LLM and agent systems.
- What do LLM and MLOps engineers earn?
- In the employer ads we read on October 11, 2026, Writer gives a base of $132K to $216K for its Software engineer, generative AI role outside San Francisco and New York. Reddit gives a base of $292,500 to $409,500 for a Senior Staff Machine Learning Engineer, GenAI Platform. On the platform side, Coinbase gives a base of $191,250 to $225,000 for a Senior Machine Learning Platform Engineer, and Samsara $200,200 to $357,500 for a Lead Machine Learning Engineer in ML Infrastructure. Every ad is linked in the table above.
- What should we test for when hiring an LLM engineer?
- Look for shipped production work, not prototypes. Ramp wants engineers who ship AI projects end to end, and Brex prefers experience with human-in-the-loop workflows, tracing, evaluation and feedback mechanisms. For a platform role, ask how they would run something like Reddit’s LLM gateway, with unified APIs, token limits and failover.
- Should LLM and MLOps engineers be on contract or permanent?
- A piece of work with a clear scope and an end date, such as a retrieval pipeline or an evaluation harness, suits an engineer on contract through their own company. A shared platform that other teams depend on, such as an LLM gateway or a feature store, needs a permanent owner. Contract to permanent lets you start one way and decide once you have seen the work.
- How soon will we see LLM or MLOps engineers?
- We agree the shortlist timing with you at the brief call, based on the role and the market. Each engineer on the shortlist is named, with our notes on the systems they have put into production.
Terms on this page
- LLM
- Large language model, the kind of model behind chat assistants and many AI products.
- MLOps
- Machine learning operations: deploying, monitoring and maintaining models in production.
- GPU
- Graphics processing unit, the chips most AI training and inference runs on.
- CUDA
- NVIDIA’s platform for programming its GPUs.
Sources
- 01
OpenAI, Applied AI Engineer, Enterprise
Dated Jul 8, 2026. Read Oct 11, 2026. jobs.ashbyhq.com
- 02
Ramp, Applied AI Engineer
Dated Jan 12, 2026. Read Oct 11, 2026. jobs.ashbyhq.com
- 03
Brex, AI Engineer, Workbench
Dated Jun 24, 2026. Read Oct 11, 2026. brex.com
- 04
Writer, Software engineer, generative AI
Dated Sep 22, 2026. Read Oct 11, 2026. jobs.ashbyhq.com
- 05
Reddit, Senior Staff Machine Learning Engineer, GenAI Platform
Dated Apr 1, 2026. Read Oct 11, 2026. job-boards.greenhouse.io
- 06
Airbnb, Principal Machine Learning Engineer- LLM Fine-tuning and Optimization
Dated May 24, 2026. Read Oct 11, 2026. careers.airbnb.com
Show all 10 sourcesShow fewer sources
- 07
Coinbase, Senior Machine Learning Platform Engineer
Dated Sep 17, 2026. Read Oct 11, 2026. coinbase.com
- 08
Chime, Senior Software Engineer, Machine Learning Platform
Dated Sep 11, 2026. Read Oct 11, 2026. boards.greenhouse.io
- 09
Gusto, Software Engineer, ML Platform
Dated Aug 10, 2026. Read Oct 11, 2026. job-boards.greenhouse.io
- 10
Samsara, Lead Machine Learning Engineer - ML Infrastructure
Dated Mar 24, 2026. Read Oct 11, 2026. samsara.com