Together AI: 23 open roles — San Francisco, USA

About Together AI

Together AI runs a cloud for training and serving open-source generative models, covering inference endpoints and GPU clusters.

People & hiring

Company scale, with the denominator visible

Employee figures and vacancy signals describe different things. Each chart names its scope and observation period.

No sourced employee headcount or technical-staff share is available for this company yet. Vacancy counts below are never used as employee estimates.

ApplyDjinn hiring evidence · current 30-day window

Explore these jobs →

Current open roles

23-1

Change versus 2026-10-07.

Technical open roles

1774%

Classified from the same current vacancies. This is a hiring mix, not the employee mix.

23

Current openings · ≤30 days

63 older active postings

2022

Founded

—

Rating

Live vacancy evidence

Technology coverage at Together AI

Coverage always uses all 23 active roles as the denominator. Scope only changes which requirement signals count.

Active roles: 23
Technologies: 40
Hard mentions: 124

AI

19 of 23 roles

83%

Core: 5%
Required: 79%
Context: 0%
Nice: 16%

Cloud

17 of 23 roles

74%

Core: 6%
Required: 59%
Context: 0%
Nice: 35%

Salesforce

14 of 23 roles

61%

Core: 0%
Required: 71%
Context: 0%
Nice: 29%

Machine Learning

12 of 23 roles

52%

Core: 0%
Required: 83%
Context: 17%
Nice: 0%

Python

7 of 23 roles

30%

Core: 0%
Required: 86%
Context: 0%
Nice: 14%

Deep Learning

6 of 23 roles

26%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

PyTorch

6 of 23 roles

26%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

NLP

4 of 23 roles

17%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Scalability

9 of 23 roles

39%

Core: 0%
Required: 33%
Context: 67%
Nice: 0%

Algorithms

7 of 23 roles

30%

Core: 0%
Required: 43%
Context: 57%
Nice: 0%

Databases

6 of 23 roles

26%

Core: 0%
Required: 50%
Context: 0%
Nice: 50%

API

5 of 23 roles

22%

Core: 0%
Required: 60%
Context: 0%
Nice: 40%

CUDA

5 of 23 roles

22%

Core: 0%
Required: 60%
Context: 0%
Nice: 40%

Go

4 of 23 roles

17%

Core: 0%
Required: 75%
Context: 0%
Nice: 25%

Kubernetes

3 of 23 roles

13%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Distributed Systems

5 of 23 roles

22%

Core: 0%
Required: 40%
Context: 0%
Nice: 60%

Performance Optimization

5 of 23 roles

22%

Core: 0%
Required: 40%
Context: 60%
Nice: 0%

Automation

3 of 23 roles

13%

Core: 0%
Required: 67%
Context: 33%
Nice: 0%

Azure

3 of 23 roles

13%

Core: 0%
Required: 67%
Context: 33%
Nice: 0%

Git

3 of 23 roles

13%

Core: 0%
Required: 67%
Context: 0%
Nice: 33%

TypeScript

3 of 23 roles

13%

Core: 0%
Required: 67%
Context: 0%
Nice: 33%

Ansible

2 of 23 roles

9%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

AWS

2 of 23 roles

9%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Backend

2 of 23 roles

9%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

GCP

2 of 23 roles

9%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Grafana

2 of 23 roles

9%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Prometheus

2 of 23 roles

9%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Reliability Engineering

5 of 23 roles

22%

Core: 0%
Required: 20%
Context: 80%
Nice: 0%

LLM

2 of 23 roles

9%

Core: 0%
Required: 50%
Context: 50%
Nice: 0%

CI/CD

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

InfiniBand

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Kafka

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Microservices

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

PCIe

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

RAG

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Software Architecture

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

Terraform

1 of 23 roles

4%

Core: 0%
Required: 100%
Context: 0%
Nice: 0%

AI Agents

3 of 23 roles

13%

Core: 0%
Required: 0%
Context: 100%
Nice: 0%

GenAI

2 of 23 roles

9%

Core: 0%
Required: 0%
Context: 100%
Nice: 0%

High Availability

1 of 23 roles

4%

Core: 0%
Required: 0%
Context: 100%
Nice: 0%

Curated company stack

Company-level metadata, kept separate from live vacancy requirements.

PythonCUDAC++PyTorchKubernetesCloudMachine Learning

Perks & Benefits

✓Open model infrastructure

✓Hybrid work

✓Equity

✓Benefits

Company Details

Industry

AI Infrastructure / Model Serving

Company size

250+

Headquarters

San Francisco, USA

Founded

2022

Domain

together.ai

Rating

—

Open roles

23

Technical roles

17 of 23

Observed hiring change

+9 roles

Hiring and market history

Daily vacancy snapshots show this company’s observed hiring; no standalone public ticker is assigned.

Open roles

Vacancy history

Current role mix

17 technical · 6 other

74% of currently observed roles are technical.

Employee perspectives

Reviews of Together AI

Search Google for employees’ reviews on Glassdoor. ApplyDjinn does not copy review text.

Glassdoor may ask you to sign in there to see more reviews.

Company monitor

Published about Together AI

Official updates and independent stories that name Together AI, with dates and source attribution. Each preview opens here first.

8 verified updatesOfficial feed monitored

View news and attention signals for Together AI →

Official · together.aiOct 6, 2026

Expanding our enterprise inference capacity with IBM Cloud and NVIDIA

Enterprises can now run open models at production scale on a dedicated B300 inference cluster, built by Together AI, IBM Cloud, and NVIDIA

Open story →

Official · together.aiOct 5, 2026

Together Link: open models in the harness you already use. Start with one command today.

Together Link brings frontier open models like GLM 5.3 and Kimi K3 into the coding agent your team already uses, cutting model spend by over 50%.

Open story →

Official · together.aiSep 23, 2026

How to train your own Jev for $17

We just launched our own Jev-like classifier, together/Tev1-4B-experimental, on top of Qwen3.5 4B on Together’s serverless platform. In this blog post we’ll show you how to fine-tune your own version!

Open story →

Official · together.aiSep 22, 2026

Canary rollouts: upgrade models in production without downtime

A hard model swap exposes every user at once, and rolling back means cold-starting the old deployment under pressure. Here's how staged traffic ramps, metric gates, and automatic rollback work on dedicated inference.

Open story →

Official · together.aiSep 18, 2026

How a global fintech scaled coding agent traffic with Dedicated Model Inference

Inside a global bank's shift to self-serve dedicated inference: how Together's DMI gave engineering teams direct control over scaling, models, and testing.

Open story →

Official · together.aiSep 16, 2026

Migrating from closed to open source models, Together

Moving from closed to open source models can take weeks, not years. A five-stage playbook: discover, evaluate, adapt, decide, and production.

Open story →

Official · together.aiSep 11, 2026

Together AI expands fine-tuning service with more models, live metrics, and finer controls

Together Fine-Tuning adds the latest open-weight models, live experiment tracking, Expert LoRA, early stopping, tokenized dataset previews, pre-flight validation, and lower training prices on selected models.

Open story →

Official · together.aiSep 10, 2026

To Infinity and Beyond: ThunderKittens Now on NVIDIA Vera Rubin NVL72!

We ported ThunderKittens to NVIDIA's Vera Rubin NVL72 and rebuilt our NVFP4 GEMM around the new hardware, taking it from 42% of roofline to over 22 PFLOPS — competitive with cuBLAS and CuTe DSL. Here is what changed in the ISA and how we used it.

Open story →