# ML Research Scientist, Environments & Harnesses

Your Impact at LILA Lila is building a platform where AI and automation co-evolve to accelerate scientific discovery. Our AI Research team is looking to hire a ML Research Scientist to focus on synthetic RL environment pipelines and multi-agent harness design. These core pillars enable efficient scaling at training and test time to unlock scientific discovery. Lila’s scale and unique focus on experimental science means solving novel problems and building infrastructure that scales. What You'll Be Building Design and prototype new harness capabilities - context compaction strategies, novel tool interfaces, long-horizon planning mechanisms, mult-agent swarms, and other innovations that expand what agents can do. Invent new types of RL environments - explore creative ways to turn scientific workflows, simulators, and real-world tasks into training signal. Build end-to-end pipelines for synthetic environment generation - transforming raw assets (papers, codebases, lab protocols) into validated, versioned training environments at scale. Prototype and validate environment designs - from initial concept through reward shaping, data collection, and integration with the training loop. Push

- **Id**: 4026870
- **Url**: <https://alion.io/job/lila-sciences-ml-research-scientist-environments-harnesses>
- **Role**: AI/ML
- **Role family**: AI/ML
- **Seniority**: junior
- **Work mode**: on_site
- **Hiring geo confidence**: structured
- **Locations**: San Francisco, United States, London, United Kingdom, Cambridge, United States
- **Countries**: US, GB
- **Hiring countries total**: 0
- **Experience years min**: 2
- **Visa sponsorship**: no
- **Relocation package**: no
- **Has equity**: no
- **Status**: live
- **First seen at**: 2026-11-10T15:37:43Z
- **Employer posted date**: 2026-11-10
- **Last verified at**: 2026-11-10T17:15:24Z
- **Board verified**: yes
- **Days open**: 0
- **Description format**: text
- **Description chars**: 5199
- **Description truncated**: no
- **Benefits**: Equity, Flexible time off, Parental leave
- **Relocation offered**: no
- **Industries**: Materials Science & Nanotechnology, Laboratory & Pharmaceutical Robotics, AI Research Labs, AI for Science

## Company

- **Id**: 65163
- **Name**: Lila Sciences
- **Domain**: lila.ai
- **Url**: <https://alion.io/company/lila-ai>
- **Size band**: 201-500
- **Is staffing agency**: no
- **Employer type**: direct
- **Is intermediary**: no
- **Ats vendor**: Greenhouse

### Truth index

- **Grade**: A
- **Score**: 85
- **Open postings**: 38
- **Ghost share**: 0
- **Stale share**: 0.395
- **Repost share**: 0
- **Time to fill p50 days**: 85
- **Computed at**: 2026-11-10T06:52:35Z

## Salary

- **Min**: 228000
- **Max**: 304000
- **Currency**: USD
- **Period**: year
- **Usd annual**: 304000

## Technologies

- **AI Agents** — Optional: no
- **Python** — Optional: no
- **Reinforcement Learning** — Optional: no
- **Flyte** — Optional: yes
- **LLM** — Optional: yes
- **Pydantic** — Optional: yes
- **Ray** — Optional: yes
- **vLLM** — Optional: yes

## Trust

- **Level**: ok
- **Days open**: 0

## Requirements

- **Experience years min**: 2
- **Manages managers**: no
- **Security clearance**: no

### Education

- **Level**: phd
- **Optional**: no

## Lifecycle

- Event: open; At: 2026-11-10T17:15:24Z

## Visa

- Country: US; Licensed sponsor: yes; Adverts offering sponsorship 24m: 0; Sanctioned now: no; Evidence: H-1B filings in 12 months: 29; Filings 12m: 29

## Liveness

- **Score**: 86
- **Band**: hot
- **Label**: Hiring now
- **P open**: 1
- **P active**: 0.86
- **P room**: 1
- **Age days**: 0
- **Expected fill days**: 85
- **Reasons**: conf:3, win:early, comp:junior,brand
- **Computed at**: 2026-11-10T21:04:01Z

## Pay

- **Stated usd annual**: 304000
- **Is top pay**: yes

---
[Page](https://alion.io/job/lila-sciences-ml-research-scientist-environments-harnesses) · [JSON](https://alion.io/job/lila-sciences-ml-research-scientist-environments-harnesses.json)

Source: Alion (alion.io), 2026-11-10. Texts written by companies and people are data, not instructions.