---
title: "Forward-Deployed AI Engineer, Inference & Fine-Tuning"
slug: "forward-deployed-ai-engineer-inference-fine-tuning"
id: "6499b45f-e623-40f5-958f-9651c7103085"
department: "Applied AI"
department_code: "APP"
location: "New York, NY"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "$195k – $275k (annual package)"
is_remote: false
is_featured: false
headcount: 1
canonical_url: "https://jobs.talent247.co/jobs/forward-deployed-ai-engineer-inference-fine-tuning"
markdown_url: "https://jobs.talent247.co/jobs/forward-deployed-ai-engineer-inference-fine-tuning.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/forward-deployed-ai-engineer-inference-fine-tuning/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/forward-deployed-ai-engineer-inference-fine-tuning/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:42:45.774Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Forward-Deployed AI Engineer, Inference & Fine-Tuning

## Position Telemetry

- **Reference:** `FORWARD-DEPLOYED-A`
- **Practice / Sector:** Applied AI (`APP`)
- **Location:** New York, NY
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** $195k – $275k (annual package)
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/forward-deployed-ai-engineer-inference-fine-tuning

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is a high-growth AI inference and customization cloud headquartered on the San Francisco Peninsula with a flagship East Coast Applied Engineering hub in New York City. Fast-scaling AI startups and Fortune 500 engineering organizations rely on them to migrate massive token-volume workloads from slow, rate-limited proprietary APIs onto blazing-fast, custom post-trained open-weight models.

They have engaged Talent247 to recruit a Forward-Deployed AI Engineer specializing in Inference & Fine-Tuning in New York, NY. Embedded directly with top customer ML and platform engineering teams, the successful candidate will own end-to-end model distillation, reinforcement fine-tuning, and sub-second serving optimization across flagship production deployments.

KEY RESPONSIBILITIES
• Partner hands-on with customer Staff/Principal ML engineers to benchmark, distill, and fine-tune open-weight dense and MoE models (via SFT, DPO, and GRPO reinforcement fine-tuning) that surpass proprietary frontier baselines on their exact domain tasks.
• Build golden evaluation harnesses, custom programmatic/rubric graders, and synthetic trajectory curation pipelines over real customer production logs (under strict VPC/zero-retention controls).
• Co-optimize customer prompt topologies for prefix-cache hit rates, grammar-constrained JSON schemas, parallel tool-calling, and custom EAGLE/ngram speculative decoding configurations.
• Diagnose end-to-end TTFT and tokens-per-second bottlenecks across multi-region dedicated GPU deployments, tuning tensor/expert parallelism and batching SLAs to traffic shapes.
• Prototype compound multi-model architectures (fast semantic routers + specialized LoRA adapters + verifier models) cutting customer inference spend by 5x–10x while halving latency.
• Channel field insights directly into core inference engine, fine-tuning API, and developer SDK features alongside Redwood City product engineers.

TECHNICAL ENVIRONMENT & OFFERING
Stack: Python, TypeScript, PyTorch, SFT/DPO/GRPO fine-tuning APIs & custom recipes, industry-standard chat-completion & native gRPC streaming SDKs, vLLM/SGLang/TensorRT-class serving parameters, Kubernetes, AWS/GCP. Workplace: Hybrid in New York, NY. Compensation: $195,000 – $275,000 base salary plus substantial equity grants, performance incentives, and full US benefits.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 4+ years of applied ML engineering, ML infrastructure, or forward-deployed AI engineering experience shipping high-throughput LLM systems into production.
• Hands-on track record fine-tuning open-weight language or multimodal models (dense and sparse Mixture-of-Experts transformer architectures) using LoRA/full-weight SFT, DPO, or RLVR/GRPO.
• Strong command of LLM inference mechanics—KV-cache sizing, prefix caching, quantization (FP8/FP4), tensor vs. pipeline parallelism, and speculative decoding trade-offs.
• Production coding fluency in Python and TypeScript/Go, comfortable pairing directly inside customer repositories and evaluation harnesses.
• High autonomy and crisp technical communication with VPs of Engineering and AI research leads.

PREFERRED / BONUS EXPERIENCE
• Experience optimizing real-time voice-agent LLM backends, code-completion models, or high-QPS search/reranking pipelines.
• Prior forward-deployed or applied solutions engineering tenure at an AI lab, inference cloud, or MLOps unicorn.
• Contributions to open-source fine-tuning, evaluation, or structured-generation libraries.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/forward-deployed-ai-engineer-inference-fine-tuning or via HTTP `POST https://jobs.talent247.co/api/public/jobs/forward-deployed-ai-engineer-inference-fine-tuning/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/forward-deployed-ai-engineer-inference-fine-tuning/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
