$60-$200/hr software engineering work, on your schedule
Review AI pull requests for the bug that compiles but is wrong, the query that kills prod at scale, the abstraction every senior would flag. Paid hourly, remote, a few hours a week.
Trusted by top research companies
Hi, we're Zac and Jack, the founders of Terac. We want to talk to you directly, because you are the most important part of what we're building.
Terac is a community of experts. People who have spent years getting good at something specific and hard. The world is about to need more of you, not less. As AI takes on more of the world's work, the bottleneck shifts to the people who actually know what they're talking about.
Expert labor is the rarest resource in the world right now, and it is shockingly hard to find. The companies that need a senior engineer's eye on a tricky code review spend weeks chasing people, paying placement fees, and settling for whoever is available. Meanwhile thousands of qualified people are sitting with knowledge that no one ever asks for.
That gap is what we're here to close. Every project that lands on Terac is routed to the people who actually know the answer, on their schedule, paid fairly, and only when the work is verified. No middleman taking a cut of your time. No vague gigs. No chasing checks.
We care about every single person in this community. If you join Terac, you're not a row in a database to us. We read the feedback. We answer the emails. We will fight for you when a customer is being unreasonable, and we will be honest with you when something on our side is broken. The quality of this panel is our entire company, and we owe you a serious bar.
If you've made it this far, here is what we're asking: claim your profile. Put your expertise on the record. Let the world's most ambitious teams come find you for the work only you can do.
Software Engineering questions
Still curious? Write to us at support@terac.com.
Niche depth is often the most valuable, because models are weakest where the expert population is small and training data is thin. Embedded C, RTOS internals, LLVM backend work, and SMT solvers are exactly those areas. You are matched to tasks in your sub-domain, not generic coding problems.
Mostly reviewing AI code for correctness, security, and idiomatic style in Rust, Go, or Python; evaluating design and architecture proposals against real constraints; and writing worked examples that show your reasoning, like debugging a race condition or sizing a cache. You never write production code for a client.
Cloud and infra certs are real signals for tasks on IAM policy, networking, or Kubernetes, where a model's answer looks plausible but hides a subtle misconfiguration. Certs are not required, but holding one prioritizes you over generalist reviewers. Lapsed certs still count as evidence of real depth.
No. Every task uses synthetic or publicly licensed code built for training, never proprietary systems or real production codebases. You are never shown confidential data, internal schemas, or anything covered by an NDA. If a scenario resembles something you recognize as confidential, flag it and it gets replaced.
Security work is directly applicable and in high demand: judging whether AI auth flows, crypto, or input validation are actually safe, not just syntactically correct. Pen-testing helps on adversarial tasks where you hunt exploitable flaws in code that passed surface review. CISSP, OSCP, and similar credentials are strong signals.
Why your expertise matters
A model produces code that compiles and looks idiomatic but kills production at 10k rows a second, or hides a footgun every senior engineer on your team would flag on sight. Catching the gap between code that runs and code you would trust takes a working engineer, not a benchmark author. Your reviews teach the model the difference.
How pay works
Top of the $60-$200 band goes to high-signal depth: systems programming, distributed correctness, security-critical review, compiler internals. Work is remote, logged hourly, and paid once your deliverable is verified. No unpaid trials, no minimum hours.
What the work looks like
A sample of the software engineering work you would pick up. Every project is scoped, remote, and paid on verified completion.
- Review a model's Go service for data races and judge whether a sync.Mutex fix is right or a channel-based approach fits the codebase better.
- Write a canonical worked example of a distributed lock with fencing tokens, annotating each design call so the model learns the tradeoffs.
- Evaluate a machine-written PostgreSQL migration on a high-traffic table for index type, ALTER TABLE locking, and a safe rollback path.
- Compare two AI-drafted binary search tree deletions and label which one handles all three deletion cases, with a written rationale.
- Stress-test a model's Rust by checking whether the lifetime annotations compile and the unsafe block upholds the safe API's invariants.
- Rate a set of AI pull request reviews on a React codebase for whether they separate real re-render issues from cosmetic bundle-size noise.
Specialties we match
Software Engineering projects span a wide range of focus areas. Tell us where you go deep and we route the work that fits.
- Distributed systems design
- Concurrency and race condition analysis
- Systems programming (C, C++, Rust)
- API design and versioning
- Security code review (OWASP, CWE)
- Compiler internals and IR
- Database query optimization
- Observability and tracing (OpenTelemetry)
- CI/CD pipeline architecture
- Type system design
- Embedded and real-time systems
- Performance profiling and benchmarking
